Menu
▾
▴
dl-learner-svn — DL-Learner Subversion commits
You can subscribe to this list here.
| 2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(120) |
Sep
(36) |
Oct
(116) |
Nov
(17) |
Dec
(44) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2008 |
Jan
(143) |
Feb
(192) |
Mar
(74) |
Apr
(84) |
May
(105) |
Jun
(64) |
Jul
(49) |
Aug
(120) |
Sep
(159) |
Oct
(156) |
Nov
(51) |
Dec
(28) |
| 2009 |
Jan
(17) |
Feb
(55) |
Mar
(33) |
Apr
(57) |
May
(54) |
Jun
(28) |
Jul
(6) |
Aug
(16) |
Sep
(38) |
Oct
(30) |
Nov
(26) |
Dec
(52) |
| 2010 |
Jan
(7) |
Feb
(91) |
Mar
(65) |
Apr
(2) |
May
(14) |
Jun
(25) |
Jul
(38) |
Aug
(48) |
Sep
(80) |
Oct
(70) |
Nov
(75) |
Dec
(77) |
| 2011 |
Jan
(68) |
Feb
(53) |
Mar
(51) |
Apr
(35) |
May
(65) |
Jun
(101) |
Jul
(29) |
Aug
(230) |
Sep
(95) |
Oct
(49) |
Nov
(110) |
Dec
(63) |
| 2012 |
Jan
(41) |
Feb
(42) |
Mar
(25) |
Apr
(46) |
May
(51) |
Jun
(44) |
Jul
(45) |
Aug
(29) |
Sep
(12) |
Oct
(9) |
Nov
(17) |
Dec
(2) |
| 2013 |
Jan
(12) |
Feb
(14) |
Mar
(7) |
Apr
(16) |
May
(54) |
Jun
(27) |
Jul
(11) |
Aug
(5) |
Sep
(85) |
Oct
(27) |
Nov
(37) |
Dec
(32) |
| 2014 |
Jan
(8) |
Feb
(29) |
Mar
(5) |
Apr
(3) |
May
(22) |
Jun
(3) |
Jul
(4) |
Aug
(3) |
Sep
|
Oct
|
Nov
|
Dec
|
|
From: <lor...@us...> - 2013-03-04 15:00:44
|
Revision: 3910
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3910&view=rev
Author: lorenz_b
Date: 2013-03-04 15:00:36 +0000 (Mon, 04 Mar 2013)
Log Message:
-----------
Changed OWL API version.
Modified Paths:
--------------
trunk/scripts/pom.xml
Modified: trunk/scripts/pom.xml
===================================================================
--- trunk/scripts/pom.xml 2013-03-04 10:48:54 UTC (rev 3909)
+++ trunk/scripts/pom.xml 2013-03-04 15:00:36 UTC (rev 3910)
@@ -116,7 +116,7 @@
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
<artifactId>owlapi-distribution</artifactId>
- <version>3.4.4-SNAPSHOT</version>
+<!-- <version>3.4.4-SNAPSHOT</version> -->
</dependency>
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-03-04 10:49:01
|
Revision: 3909
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3909&view=rev
Author: lorenz_b
Date: 2013-03-04 10:48:54 +0000 (Mon, 04 Mar 2013)
Log Message:
-----------
Updated SOLR lib.
Modified Paths:
--------------
trunk/components-ext/src/main/java/org/dllearner/common/index/SOLRIndex.java
trunk/pom.xml
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/SOLRIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/SOLRIndex.java 2013-03-02 09:41:31 UTC (rev 3908)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/SOLRIndex.java 2013-03-04 10:48:54 UTC (rev 3909)
@@ -1,6 +1,5 @@
package org.dllearner.common.index;
-import java.net.MalformedURLException;
import java.util.ArrayList;
import java.util.List;
@@ -8,7 +7,7 @@
import org.apache.solr.client.solrj.SolrQuery.ORDER;
import org.apache.solr.client.solrj.SolrServerException;
import org.apache.solr.client.solrj.impl.BinaryRequestWriter;
-import org.apache.solr.client.solrj.impl.CommonsHttpSolrServer;
+import org.apache.solr.client.solrj.impl.HttpSolrServer;
import org.apache.solr.client.solrj.response.QueryResponse;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.common.SolrDocumentList;
@@ -16,7 +15,7 @@
public class SOLRIndex implements Index{
-private CommonsHttpSolrServer server;
+private HttpSolrServer server;
private static final int DEFAULT_LIMIT = 10;
private static final int DEFAULT_OFFSET = 0;
@@ -29,12 +28,8 @@
private boolean restrictiveSearch = true;
public SOLRIndex(String solrServerURL){
- try {
- server = new CommonsHttpSolrServer(solrServerURL);
- server.setRequestWriter(new BinaryRequestWriter());
- } catch (MalformedURLException e) {
- e.printStackTrace();
- }
+ server = new HttpSolrServer(solrServerURL);
+ server.setRequestWriter(new BinaryRequestWriter());
}
public void setSearchFields(String primarySearchField, String secondarySearchField){
Modified: trunk/pom.xml
===================================================================
--- trunk/pom.xml 2013-03-02 09:41:31 UTC (rev 3908)
+++ trunk/pom.xml 2013-03-04 10:48:54 UTC (rev 3909)
@@ -20,6 +20,7 @@
<slf4j.version>1.6.4</slf4j.version>
<log4j.version>1.2.16</log4j.version>
+ <solr.version>4.1.0</solr.version>
</properties>
<modules>
@@ -164,7 +165,7 @@
<dependency>
<groupId>org.apache.solr</groupId>
<artifactId>solr-core</artifactId>
- <version>3.5.0</version>
+ <version>${solr.version}</version>
<exclusions>
<exclusion>
<groupId>commons-logging</groupId>
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <dc...@us...> - 2013-03-02 09:41:38
|
Revision: 3908
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3908&view=rev
Author: dcherix
Date: 2013-03-02 09:41:31 +0000 (Sat, 02 Mar 2013)
Log Message:
-----------
Removed magic strings from scripts
Modified Paths:
--------------
trunk/scripts/src/main/java/org/dllearner/scripts/improveWikipedia/DBpediaClassLearnerCELOE.java
trunk/scripts/src/main/java/org/dllearner/scripts/improveWikipedia/NewSparqlCompDBpediaClassLearnerCELOE.java
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/improveWikipedia/DBpediaClassLearnerCELOE.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/improveWikipedia/DBpediaClassLearnerCELOE.java 2013-02-27 15:25:49 UTC (rev 3907)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/improveWikipedia/DBpediaClassLearnerCELOE.java 2013-03-02 09:41:31 UTC (rev 3908)
@@ -453,4 +453,4 @@
}
}
-}
+}
\ No newline at end of file
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/improveWikipedia/NewSparqlCompDBpediaClassLearnerCELOE.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/improveWikipedia/NewSparqlCompDBpediaClassLearnerCELOE.java 2013-02-27 15:25:49 UTC (rev 3907)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/improveWikipedia/NewSparqlCompDBpediaClassLearnerCELOE.java 2013-03-02 09:41:31 UTC (rev 3908)
@@ -433,4 +433,4 @@
return results;
}
-}
+}
\ No newline at end of file
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-27 15:25:56
|
Revision: 3907
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3907&view=rev
Author: lorenz_b
Date: 2013-02-27 15:25:49 +0000 (Wed, 27 Feb 2013)
Log Message:
-----------
Small fix.
Modified Paths:
--------------
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-02-27 13:48:57 UTC (rev 3906)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-02-27 15:25:49 UTC (rev 3907)
@@ -19,18 +19,15 @@
import org.apache.log4j.Logger;
import org.dllearner.algorithms.celoe.CELOE;
-import org.dllearner.algorithms.qtl.QTL;
import org.dllearner.core.AbstractLearningProblem;
import org.dllearner.core.AbstractReasonerComponent;
import org.dllearner.core.ComponentInitException;
import org.dllearner.core.EvaluatedDescription;
import org.dllearner.core.KnowledgeSource;
-import org.dllearner.core.LearningProblemUnsupportedException;
import org.dllearner.core.owl.Description;
import org.dllearner.core.owl.Individual;
import org.dllearner.core.owl.NamedClass;
import org.dllearner.core.owl.ObjectProperty;
-import org.dllearner.kb.LocalModelBasedSparqlEndpointKS;
import org.dllearner.kb.OWLAPIOntology;
import org.dllearner.kb.SparqlEndpointKS;
import org.dllearner.kb.sparql.ConciseBoundedDescriptionGenerator;
@@ -44,12 +41,10 @@
import org.dllearner.learningproblems.PosOnlyLP;
import org.dllearner.reasoning.FastInstanceChecker;
import org.dllearner.reasoning.SPARQLReasoner;
-import org.dllearner.utilities.CrossValidation;
import org.dllearner.utilities.LabelShortFormProvider;
import org.dllearner.utilities.datastructures.Datastructures;
import org.dllearner.utilities.datastructures.SetManipulation;
import org.dllearner.utilities.datastructures.SortedSetTuple;
-import org.dllearner.utilities.examples.AutomaticNegativeExampleFinderSPARQL2;
import org.dllearner.utilities.owl.OWLAPIDescriptionConvertVisitor;
import org.semanticweb.owlapi.apibinding.OWLManager;
import org.semanticweb.owlapi.model.OWLOntology;
@@ -320,7 +315,7 @@
logger.info("Done.");
if(performCrossValidation){
- CrossValidation cv = new CrossValidation(la, lp, rc, 5, false);
+ org.dllearner.cli.CrossValidation cv = new org.dllearner.cli.CrossValidation(la, lp, rc, 5, false);
} else {
//apply the learning algorithm
logger.info("Running learning algorithm...");
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-27 13:49:04
|
Revision: 3906
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3906&view=rev
Author: lorenz_b
Date: 2013-02-27 13:48:57 +0000 (Wed, 27 Feb 2013)
Log Message:
-----------
Added cross validation to matching tests.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/kb/sparql/SparqlEndpoint.java
trunk/interfaces/src/main/java/org/dllearner/cli/CrossValidation.java
trunk/scripts/pom.xml
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java
Modified: trunk/components-core/src/main/java/org/dllearner/kb/sparql/SparqlEndpoint.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/kb/sparql/SparqlEndpoint.java 2013-02-25 12:24:29 UTC (rev 3905)
+++ trunk/components-core/src/main/java/org/dllearner/kb/sparql/SparqlEndpoint.java 2013-02-27 13:48:57 UTC (rev 3906)
@@ -210,6 +210,18 @@
return new SparqlEndpoint(u, defaultGraphURIs, new LinkedList<String>());
}
+ public static SparqlEndpoint getEndpointLOD2Cloud() {
+ URL u = null;
+ try {
+ u = new URL("http://lod.openlinksw.com/sparql/");
+ } catch (Exception e) {
+ e.printStackTrace();
+ }
+ LinkedList<String> defaultGraphURIs=new LinkedList<String>();
+// defaultGraphURIs.add("http://dbpedia.org");
+ return new SparqlEndpoint(u, defaultGraphURIs, new LinkedList<String>());
+ }
+
public static SparqlEndpoint getEndpointLinkedGeoData() {
URL u = null;
try {
Modified: trunk/interfaces/src/main/java/org/dllearner/cli/CrossValidation.java
===================================================================
--- trunk/interfaces/src/main/java/org/dllearner/cli/CrossValidation.java 2013-02-25 12:24:29 UTC (rev 3905)
+++ trunk/interfaces/src/main/java/org/dllearner/cli/CrossValidation.java 2013-02-27 13:48:57 UTC (rev 3906)
@@ -27,7 +27,9 @@
import java.util.List;
import java.util.Random;
import java.util.Set;
+import java.util.TreeSet;
+import org.dllearner.core.AbstractLearningProblem;
import org.dllearner.core.ComponentInitException;
import org.dllearner.core.AbstractCELA;
import org.dllearner.core.AbstractReasonerComponent;
@@ -35,6 +37,7 @@
import org.dllearner.core.owl.Individual;
import org.dllearner.learningproblems.Heuristics;
import org.dllearner.learningproblems.PosNegLP;
+import org.dllearner.learningproblems.PosOnlyLP;
import org.dllearner.utilities.Helper;
import org.dllearner.utilities.datastructures.Datastructures;
import org.dllearner.utilities.statistics.Stat;
@@ -70,7 +73,7 @@
}
- public CrossValidation(AbstractCELA la, PosNegLP lp, AbstractReasonerComponent rs, int folds, boolean leaveOneOut) {
+ public CrossValidation(AbstractCELA la, AbstractLearningProblem lp, AbstractReasonerComponent rs, int folds, boolean leaveOneOut) {
DecimalFormat df = new DecimalFormat();
@@ -81,11 +84,20 @@
List<Set<Individual>> testSetsNeg = new LinkedList<Set<Individual>>();
// get examples and shuffle them too
- Set<Individual> posExamples = ((PosNegLP)lp).getPositiveExamples();
+ Set<Individual> posExamples;
+ Set<Individual> negExamples;
+ if(lp instanceof PosNegLP){
+ posExamples = ((PosNegLP)lp).getPositiveExamples();
+ negExamples = ((PosNegLP)lp).getNegativeExamples();
+ } else if(lp instanceof PosOnlyLP){
+ posExamples = ((PosNegLP)lp).getPositiveExamples();
+ negExamples = new HashSet<Individual>();
+ } else {
+ throw new IllegalArgumentException("Only PosNeg and PosOnly learning problems are supported");
+ }
List<Individual> posExamplesList = new LinkedList<Individual>(posExamples);
+ List<Individual> negExamplesList = new LinkedList<Individual>(negExamples);
Collections.shuffle(posExamplesList, new Random(1));
- Set<Individual> negExamples = ((PosNegLP)lp).getNegativeExamples();
- List<Individual> negExamplesList = new LinkedList<Individual>(negExamples);
Collections.shuffle(negExamplesList, new Random(2));
// sanity check whether nr. of folds makes sense for this benchmark
@@ -137,8 +149,13 @@
Set<String> pos = Datastructures.individualSetToStringSet(trainingSetsPos.get(currFold));
Set<String> neg = Datastructures.individualSetToStringSet(trainingSetsNeg.get(currFold));
- lp.setPositiveExamples(trainingSetsPos.get(currFold));
- lp.setNegativeExamples(trainingSetsNeg.get(currFold));
+ if(lp instanceof PosNegLP){
+ ((PosNegLP)lp).setPositiveExamples(trainingSetsPos.get(currFold));
+ ((PosNegLP)lp).setNegativeExamples(trainingSetsNeg.get(currFold));
+ } else if(lp instanceof PosOnlyLP){
+ ((PosOnlyLP)lp).setPositiveExamples(new TreeSet<Individual>(trainingSetsPos.get(currFold)));
+ }
+
try {
lp.init();
Modified: trunk/scripts/pom.xml
===================================================================
--- trunk/scripts/pom.xml 2013-02-25 12:24:29 UTC (rev 3905)
+++ trunk/scripts/pom.xml 2013-02-27 13:48:57 UTC (rev 3906)
@@ -116,7 +116,7 @@
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
<artifactId>owlapi-distribution</artifactId>
- <version>3.4</version>
+ <version>3.4.4-SNAPSHOT</version>
</dependency>
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-02-25 12:24:29 UTC (rev 3905)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-02-27 13:48:57 UTC (rev 3906)
@@ -44,6 +44,7 @@
import org.dllearner.learningproblems.PosOnlyLP;
import org.dllearner.reasoning.FastInstanceChecker;
import org.dllearner.reasoning.SPARQLReasoner;
+import org.dllearner.utilities.CrossValidation;
import org.dllearner.utilities.LabelShortFormProvider;
import org.dllearner.utilities.datastructures.Datastructures;
import org.dllearner.utilities.datastructures.SetManipulation;
@@ -75,6 +76,8 @@
import com.hp.hpl.jena.rdf.model.Statement;
import com.hp.hpl.jena.sparql.engine.http.QueryEngineHTTP;
import com.hp.hpl.jena.vocabulary.OWL;
+import com.hp.hpl.jena.vocabulary.RDF;
+import com.hp.hpl.jena.vocabulary.RDFS;
import com.jamonapi.Monitor;
import com.jamonapi.MonitorFactory;
@@ -95,6 +98,8 @@
private Map<Description, List<? extends EvaluatedDescription>> mappingKB2KB1;
private boolean posNegLearning = true;
+ private final boolean performCrossValidation = true;
+ private int fragmentDepth = 2;
/**
* The maximum number of positive examples, used for the SPARQL extraction and learning algorithm
@@ -118,6 +123,10 @@
this(new KnowledgeBase(endpoint1), new KnowledgeBase(endpoint2));
}
+ public void setFragmentDepth(int fragmentDepth) {
+ this.fragmentDepth = fragmentDepth;
+ }
+
public void start(){
mappingKB1KB2 = computeAlignment(kb1, kb2);
printMappingPretty(mappingKB1KB2);
@@ -283,14 +292,16 @@
fullFragment.add(positiveFragment);
fullFragment.add(negativeFragment);
- //here is the most difficult task, i.e. find a 'good' fragment of the KB on which we can learn
KnowledgeSource ks = convert(fullFragment);
//initialize the reasoner
+ logger.info("Initializing reasoner...");
AbstractReasonerComponent rc = new FastInstanceChecker(ks);
rc.init();
+ logger.info("Done.");
//initialize the learning problem
+ logger.info("Initializing learning problem...");
AbstractLearningProblem lp;
if(posNeg){
lp = new PosNegLPStandard(rc, positiveExamplesSample, negativeExamplesSample);
@@ -298,26 +309,34 @@
lp = new PosOnlyLP(rc, positiveExamplesSample);
}
lp.init();
+ logger.info("Done.");
- //apply the learning algorithm
- logger.info("Running learning algorithm...");
+ //initialize the learning algorithm
+ logger.info("Initializing learning algorithm...");
CELOE la = new CELOE(lp, rc);
la.setMaxExecutionTimeInSeconds(10);
la.setNoisePercentage(25);
la.init();
- la.start();
+ logger.info("Done.");
- try {
- QTL qtl = new QTL(lp, new LocalModelBasedSparqlEndpointKS(fullFragment));
- qtl.init();
- qtl.start();
- System.out.println(qtl.getSPARQLQuery());
- } catch (LearningProblemUnsupportedException e) {
- e.printStackTrace();
+ if(performCrossValidation){
+ CrossValidation cv = new CrossValidation(la, lp, rc, 5, false);
+ } else {
+ //apply the learning algorithm
+ logger.info("Running learning algorithm...");
+ la.start();
+ logger.info(la.getCurrentlyBestEvaluatedDescription());
}
-
- logger.info(la.getCurrentlyBestEvaluatedDescription());
+// try {
+// QTL qtl = new QTL(lp, new LocalModelBasedSparqlEndpointKS(fullFragment));
+// qtl.init();
+// qtl.start();
+// System.out.println(qtl.getSPARQLQuery());
+// } catch (LearningProblemUnsupportedException e) {
+// e.printStackTrace();
+// }
+
return la.getCurrentlyBestEvaluatedDescriptions(10);
} catch (ComponentInitException e) {
e.printStackTrace();
@@ -382,21 +401,32 @@
logger.info(i++ + "/" + size);
fullFragment.add(getFragment(ind, kb));
}
- //filter out triples with String literals, as there often occur are some syntax errors and they are not relevant for learning
+ filter(fullFragment);
+ return fullFragment;
+ }
+
+ private void filter(Model model) {
+ // filter out triples with String literals, as there often occur are
+ // some syntax errors and they are not relevant for learning
List<Statement> statementsToRemove = new ArrayList<Statement>();
- for(Iterator<Statement> iter = fullFragment.listStatements().toList().iterator(); iter.hasNext();){
+ for (Iterator<Statement> iter = model.listStatements().toList().iterator(); iter.hasNext();) {
Statement st = iter.next();
RDFNode object = st.getObject();
- if(object.isLiteral()){
-// statementsToRemove.add(st);
+ if (object.isLiteral()) {
+ // statementsToRemove.add(st);
Literal lit = object.asLiteral();
- if(lit.getDatatype() == null || lit.getDatatype().equals(XSD.STRING)){
+ if (lit.getDatatype() == null || lit.getDatatype().equals(XSD.STRING)) {
st.changeObject("shortened", "en");
- }
+ }
}
+ //remove statements like <x a owl:Class>
+ if(st.getPredicate().equals(RDF.type)){
+ if(object.equals(RDFS.Class.asNode()) || object.equals(OWL.Class.asNode()) || object.equals(RDFS.Literal.asNode())){
+ statementsToRemove.add(st);
+ }
+ }
}
- fullFragment.remove(statementsToRemove);
- return fullFragment;
+ model.remove(statementsToRemove);
}
/**
@@ -414,7 +444,7 @@
private Model getFragment(Individual ind, KnowledgeBase kb){
logger.debug("Loading fragment for " + ind.getName());
ConciseBoundedDescriptionGenerator cbdGen = new ConciseBoundedDescriptionGeneratorImpl(kb.getEndpoint(), kb.getCache());
- Model cbd = cbdGen.getConciseBoundedDescription(ind.getName(), 2);
+ Model cbd = cbdGen.getConciseBoundedDescription(ind.getName(), fragmentDepth);
logger.debug("Got " + cbd.size() + " triples.");
return cbd;
}
Modified: trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java
===================================================================
--- trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java 2013-02-25 12:24:29 UTC (rev 3905)
+++ trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java 2013-02-27 13:48:57 UTC (rev 3906)
@@ -33,6 +33,8 @@
private KnowledgeBase worldFactBook;
private KnowledgeBase openCyc;
private KnowledgeBase linkedGeoData;
+
+ private final int fragmentDepth = 3;
@Before
public void setUp() throws Exception {
@@ -112,6 +114,7 @@
@Test
public void testSingleClassLinkedGeoDataToDBpedia() {
OntologyMatching matcher = new OntologyMatching(linkedGeoData, dbpedia);
+ matcher.setFragmentDepth(fragmentDepth);
NamedClass nc = new NamedClass("http://linkedgeodata.org/ontology/Aerodrome");
List<? extends EvaluatedDescription> mapping = matcher.computeMapping(nc, linkedGeoData, dbpedia);
Map<Description, List<? extends EvaluatedDescription>> alignment = new HashMap<Description, List<? extends EvaluatedDescription>>();
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-25 12:24:35
|
Revision: 3905
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3905&view=rev
Author: lorenz_b
Date: 2013-02-25 12:24:29 +0000 (Mon, 25 Feb 2013)
Log Message:
-----------
Changed OWL API version.
Modified Paths:
--------------
trunk/scripts/pom.xml
Modified: trunk/scripts/pom.xml
===================================================================
--- trunk/scripts/pom.xml 2013-02-25 11:56:42 UTC (rev 3904)
+++ trunk/scripts/pom.xml 2013-02-25 12:24:29 UTC (rev 3905)
@@ -116,7 +116,7 @@
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
<artifactId>owlapi-distribution</artifactId>
- <version>3.4.4-SNAPSHOT</version>
+ <version>3.4</version>
</dependency>
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-25 11:56:48
|
Revision: 3904
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3904&view=rev
Author: lorenz_b
Date: 2013-02-25 11:56:42 +0000 (Mon, 25 Feb 2013)
Log Message:
-----------
Cleaned up POM.
Modified Paths:
--------------
trunk/components-ext/pom.xml
Modified: trunk/components-ext/pom.xml
===================================================================
--- trunk/components-ext/pom.xml 2013-02-25 11:55:05 UTC (rev 3903)
+++ trunk/components-ext/pom.xml 2013-02-25 11:56:42 UTC (rev 3904)
@@ -91,17 +91,6 @@
<!--END Logging Dependencies-->
- <dependency>
- <groupId>edu.stanford.nlp</groupId>
- <artifactId>stanford-corenlp</artifactId>
- <version>1.3.3</version>
- </dependency>
- <dependency>
- <groupId>edu.stanford.nlp</groupId>
- <artifactId>stanford-corenlp</artifactId>
- <version>1.3.3</version>
- <classifier>models</classifier>
- </dependency>
<dependency>
<groupId>lbj</groupId>
<artifactId>library</artifactId>
@@ -123,11 +112,6 @@
<version>1.0</version>
</dependency>
<dependency>
- <groupId>uk.ac.shef.wit</groupId>
- <artifactId>simmetrics</artifactId>
- <version>1.6.2</version>
- </dependency>
- <dependency>
<groupId>stax</groupId>
<artifactId>stax-api</artifactId>
<version>1.0.1</version>
@@ -138,42 +122,10 @@
<scope>test</scope>
</dependency>
<dependency>
- <groupId>org.apache.opennlp</groupId>
- <artifactId>opennlp-tools</artifactId>
- <version>1.5.1-incubating</version>
- </dependency>
- <dependency>
- <groupId>org.apache.opennlp</groupId>
- <artifactId>opennlp-maxent</artifactId>
- <version>3.0.1-incubating</version>
- </dependency>
- <dependency>
- <groupId>com.aliasi</groupId>
- <artifactId>lingpipe</artifactId>
- <version>4.0.1</version>
- </dependency>
- <dependency>
- <groupId>org.annolab.tt4j</groupId>
- <artifactId>org.annolab.tt4j</artifactId>
- <version>1.0.16</version>
- </dependency>
- <dependency>
<groupId>org.ini4j</groupId>
<artifactId>ini4j</artifactId>
</dependency>
<dependency>
- <groupId>net.didion.jwnl</groupId>
- <artifactId>jwnl</artifactId>
- <version>1.4.1.RC2</version>
- </dependency>
- <!--
- <dependency>
- <groupId>org.nlp2rdf</groupId>
- <artifactId>nif</artifactId>
- <version>1.1</version>
- </dependency>
- -->
- <dependency>
<groupId>org.xerial</groupId>
<artifactId>sqlite-jdbc</artifactId>
<version>3.7.2</version>
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-25 11:55:12
|
Revision: 3903
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3903&view=rev
Author: lorenz_b
Date: 2013-02-25 11:55:05 +0000 (Mon, 25 Feb 2013)
Log Message:
-----------
Removed TBSL algorithm.
Removed Paths:
-------------
trunk/components-ext/src/main/java/org/dllearner/algorithm/
trunk/components-ext/src/main/java/org/dllearner/common/index/IndexResultItemComparator.java
trunk/components-ext/src/test/java/org/dllearner/algorithm/
Deleted: trunk/components-ext/src/main/java/org/dllearner/common/index/IndexResultItemComparator.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/IndexResultItemComparator.java 2013-02-25 11:47:56 UTC (rev 3902)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/IndexResultItemComparator.java 2013-02-25 11:55:05 UTC (rev 3903)
@@ -1,48 +0,0 @@
-package org.dllearner.common.index;
-
-import java.util.Comparator;
-import java.util.HashMap;
-import java.util.Map;
-
-import org.dllearner.algorithm.tbsl.util.Similarity;
-
-public class IndexResultItemComparator implements Comparator<IndexResultItem>{
- private String s;
- private Map<String, Double> cache;
-
- public IndexResultItemComparator(String s) {
- this.s = s;
- cache = new HashMap<String, Double>();
- }
-
- @Override
- public int compare(IndexResultItem item1, IndexResultItem item2) {
-
- double sim1 = 0;
- if(cache.containsKey(item1.getLabel())){
- sim1 = cache.get(item1.getLabel());
- } else {
- sim1 = Similarity.getSimilarity(s, item1.getLabel());
- cache.put(item1.getLabel(), sim1);
- }
- double sim2 = 0;
- if(cache.containsKey(item2.getLabel())){
- sim2 = cache.get(item2.getLabel());
- } else {
- sim2 = Similarity.getSimilarity(s, item2.getLabel());
- cache.put(item2.getLabel(), sim2);
- }
-
- if(sim1 < sim2){
- return 1;
- } else if(sim1 > sim2){
- return -1;
- } else {
- int val = item1.getLabel().compareTo(item2.getLabel());
- if(val == 0){
- return item1.getUri().compareTo(item2.getUri());
- }
- return val;
- }
- }
-}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-25 11:48:03
|
Revision: 3902
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3902&view=rev
Author: lorenz_b
Date: 2013-02-25 11:47:56 +0000 (Mon, 25 Feb 2013)
Log Message:
-----------
Updated Staford model loading.
Modified Paths:
--------------
trunk/components-ext/pom.xml
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/nlp/StanfordLemmatizer.java
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/nlp/StanfordPartOfSpeechTagger.java
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/SPARQLEndpointMetrics.java
trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/TBSLTest.java
Modified: trunk/components-ext/pom.xml
===================================================================
--- trunk/components-ext/pom.xml 2013-02-18 14:16:54 UTC (rev 3901)
+++ trunk/components-ext/pom.xml 2013-02-25 11:47:56 UTC (rev 3902)
@@ -91,12 +91,18 @@
<!--END Logging Dependencies-->
+ <dependency>
+ <groupId>edu.stanford.nlp</groupId>
+ <artifactId>stanford-corenlp</artifactId>
+ <version>1.3.3</version>
+ </dependency>
+ <dependency>
+ <groupId>edu.stanford.nlp</groupId>
+ <artifactId>stanford-corenlp</artifactId>
+ <version>1.3.3</version>
+ <classifier>models</classifier>
+ </dependency>
<dependency>
- <groupId>edu.stanford</groupId>
- <artifactId>postagger</artifactId>
- <version>3.0.2</version>
- </dependency>
- <dependency>
<groupId>lbj</groupId>
<artifactId>library</artifactId>
<version>1.0</version>
Modified: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/nlp/StanfordLemmatizer.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/nlp/StanfordLemmatizer.java 2013-02-18 14:16:54 UTC (rev 3901)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/nlp/StanfordLemmatizer.java 2013-02-25 11:47:56 UTC (rev 3902)
@@ -26,7 +26,7 @@
@Override
public String stem(String word, String tag) {
- return stemmer.stem(word, tag).word();
+ return stemmer.lemma(word, tag);
}
@Override
Modified: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/nlp/StanfordPartOfSpeechTagger.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/nlp/StanfordPartOfSpeechTagger.java 2013-02-18 14:16:54 UTC (rev 3901)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/nlp/StanfordPartOfSpeechTagger.java 2013-02-25 11:47:56 UTC (rev 3902)
@@ -1,36 +1,29 @@
package org.dllearner.algorithm.tbsl.nlp;
-import java.io.IOException;
-import java.io.StringReader;
-import java.net.URISyntaxException;
import java.util.ArrayList;
-import java.util.Arrays;
import java.util.Collections;
import java.util.List;
+import java.util.Properties;
import com.aliasi.tag.Tagging;
-import edu.stanford.nlp.ling.HasWord;
-import edu.stanford.nlp.ling.TaggedWord;
-import edu.stanford.nlp.tagger.maxent.MaxentTagger;
+import edu.stanford.nlp.ling.CoreAnnotations.PartOfSpeechAnnotation;
+import edu.stanford.nlp.ling.CoreAnnotations.SentencesAnnotation;
+import edu.stanford.nlp.ling.CoreAnnotations.TextAnnotation;
+import edu.stanford.nlp.ling.CoreAnnotations.TokensAnnotation;
+import edu.stanford.nlp.ling.CoreLabel;
+import edu.stanford.nlp.pipeline.Annotation;
+import edu.stanford.nlp.pipeline.StanfordCoreNLP;
+import edu.stanford.nlp.util.CoreMap;
public class StanfordPartOfSpeechTagger implements PartOfSpeechTagger{
- private static final String MODEL = "tbsl/models/bidirectional-distsim-wsj-0-18.tagger";
+ private StanfordCoreNLP pipeline;
- private MaxentTagger tagger;
-
public StanfordPartOfSpeechTagger(){
- try {
-// String modelPath = this.getClass().getClassLoader().getResource(MODEL).getPath();
- String modelPath = getClass().getResource("/tbsl/models/bidirectional-distsim-wsj-0-18.tagger").getPath();
-// String modelPath = Thread.currentThread().getContextClassLoader().getResource(MODEL).getFile();
- tagger = new MaxentTagger(modelPath);
- } catch (IOException e) {
- e.printStackTrace();
- } catch (ClassNotFoundException e) {
- e.printStackTrace();
- }
+ Properties props = new Properties();
+ props.put("annotators", "tokenize, ssplit, pos");
+ pipeline = new StanfordCoreNLP(props);
}
@Override
@@ -39,68 +32,94 @@
}
@Override
- public String tag(String sentence) {
+ public String tag(String text) {
String out = "";
- ArrayList<TaggedWord> tagged = new ArrayList<TaggedWord>();
+ // create an empty Annotation just with the given text
+ Annotation document = new Annotation(text);
+
+ // run all Annotators on this text
+ pipeline.annotate(document);
+
+ // these are all the sentences in this document
+ // a CoreMap is essentially a Map that uses class objects as keys and has values with custom types
+ List<CoreMap> sentences = document.get(SentencesAnnotation.class);
+
+ for(CoreMap sentence: sentences) {
+ for (CoreLabel token: sentence.get(TokensAnnotation.class)) {
+ // this is the text of the token
+ String word = token.get(TextAnnotation.class);
+ // this is the POS tag of the token
+ String pos = token.get(PartOfSpeechAnnotation.class);
+
+ out += " " + word + "/" + pos;
+ }
+ }
- StringReader reader = new StringReader(sentence);
- List<List<HasWord>> text = MaxentTagger.tokenizeText(reader);
-
- if (text.size() == 1) {
- tagged = tagger.tagSentence(text.get(0));
- }
-
- for (TaggedWord t : tagged) {
- out += " " + t.toString();
- }
return out.trim();
}
+
+
@Override
public List<String> tagTopK(String sentence) {
return Collections.singletonList(tag(sentence));
}
- public List<String> getTags(String sentence){
+ public List<String> getTags(String text){
List<String> tags = new ArrayList<String>();
- ArrayList<TaggedWord> tagged = new ArrayList<TaggedWord>();
+ // create an empty Annotation just with the given text
+ Annotation document = new Annotation(text);
+
+ // run all Annotators on this text
+ pipeline.annotate(document);
+
+ // these are all the sentences in this document
+ // a CoreMap is essentially a Map that uses class objects as keys and has values with custom types
+ List<CoreMap> sentences = document.get(SentencesAnnotation.class);
+
+ for(CoreMap sentence: sentences) {
+ for (CoreLabel token: sentence.get(TokensAnnotation.class)) {
+ // this is the text of the token
+ String word = token.get(TextAnnotation.class);
+ // this is the POS tag of the token
+ String pos = token.get(PartOfSpeechAnnotation.class);
+
+ tags.add(pos);
+ }
+ }
- StringReader reader = new StringReader(sentence);
- List<List<HasWord>> text = MaxentTagger.tokenizeText(reader);
-
- if (text.size() == 1) {
- tagged = tagger.tagSentence(text.get(0));
- }
-
- for(TaggedWord tW : tagged){
- tags.add(tW.tag());
- }
-
return tags;
}
@Override
- public Tagging<String> getTagging(String sentence){
- ArrayList<TaggedWord> tagged = new ArrayList<TaggedWord>();
-
- StringReader reader = new StringReader(sentence);
- List<List<HasWord>> text = MaxentTagger.tokenizeText(reader);
-
- if (text.size() == 1) {
- tagged = tagger.tagSentence(text.get(0));
- }
-
+ public Tagging<String> getTagging(String text){
List<String> tokenList = new ArrayList<String>();
List<String> tagList = new ArrayList<String>();
- for(TaggedWord tW : tagged){
- tokenList.add(tW.word());
- tagList.add(tW.tag());
- }
+ // create an empty Annotation just with the given text
+ Annotation document = new Annotation(text);
+
+ // run all Annotators on this text
+ pipeline.annotate(document);
+
+ // these are all the sentences in this document
+ // a CoreMap is essentially a Map that uses class objects as keys and has values with custom types
+ List<CoreMap> sentences = document.get(SentencesAnnotation.class);
+
+ for(CoreMap sentence: sentences) {
+ for (CoreLabel token: sentence.get(TokensAnnotation.class)) {
+ // this is the text of the token
+ String word = token.get(TextAnnotation.class);
+ // this is the POS tag of the token
+ String pos = token.get(PartOfSpeechAnnotation.class);
+
+ tokenList.add(word);
+ tagList.add(pos);
+ }
+ }
return new Tagging<String>(tokenList, tagList);
}
-
}
Modified: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/SPARQLEndpointMetrics.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/SPARQLEndpointMetrics.java 2013-02-18 14:16:54 UTC (rev 3901)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/SPARQLEndpointMetrics.java 2013-02-25 11:47:56 UTC (rev 3902)
@@ -9,6 +9,7 @@
import java.util.SortedSet;
import java.util.TreeSet;
+import org.apache.log4j.Level;
import org.apache.log4j.Logger;
import org.dllearner.core.owl.Individual;
import org.dllearner.core.owl.NamedClass;
@@ -34,8 +35,8 @@
public SPARQLEndpointMetrics(SparqlEndpoint endpoint, ExtractionDBCache cache) {
this.endpoint = endpoint;
this.cache = cache;
- cache.setFreshnessInMilliseconds(31536000000l);
- cache.setMaxExecutionTimeInSeconds(30);
+ cache.setFreshnessInMilliseconds(Long.MAX_VALUE);//31536000000l);
+ cache.setMaxExecutionTimeInSeconds(300);
this.reasoner = new SPARQLReasoner(new SparqlEndpointKS(endpoint), cache);
}
@@ -214,6 +215,32 @@
}
/**
+ * Returns the number of triples where the given individual is in subject position(out-going links).
+ * @param cls
+ * @return
+ */
+ public int getOccurencesInSubjectPosition(Individual ind){
+ log.trace(String.format("Computing number of occurences in subject position for %s", ind.getName()));
+ String query = String.format("SELECT (COUNT(*) AS ?cnt) WHERE {<%s> ?p ?o.}", ind.getName());
+ ResultSet rs = SparqlQuery.convertJSONtoResultSet(cache.executeSelectQuery(endpoint, query));
+ int classOccurenceCnt = rs.next().getLiteral("cnt").getInt();
+ return classOccurenceCnt;
+ }
+
+ /**
+ * Returns the number of triples where the given individual is in object position(in-going links).
+ * @param cls
+ * @return
+ */
+ public int getOccurencesInObjectPosition(Individual ind){
+ log.trace(String.format("Computing number of occurences in object position for %s", ind.getName()));
+ String query = String.format("SELECT (COUNT(*) AS ?cnt) WHERE {?s ?p <%s>.}", ind.getName());
+ ResultSet rs = SparqlQuery.convertJSONtoResultSet(cache.executeSelectQuery(endpoint, query));
+ int classOccurenceCnt = rs.next().getLiteral("cnt").getInt();
+ return classOccurenceCnt;
+ }
+
+ /**
* Returns the number triples with the given property as predicate.
* @param prop
* @return
@@ -394,8 +421,9 @@
}
public static void main(String[] args) {
- SparqlEndpoint endpoint = SparqlEndpoint.getEndpointDBpedia();
- ExtractionDBCache cache = new ExtractionDBCache("/opt/tbsl/dbpedia_pmi_cache");
+ Logger.getLogger(SPARQLEndpointMetrics.class).setLevel(Level.DEBUG);
+ SparqlEndpoint endpoint = SparqlEndpoint.getEndpointDBpediaLiveOpenLink();
+ ExtractionDBCache cache = new ExtractionDBCache("/opt/tbsl/dbpedia_pmi_cache_v2");
String NS = "http://dbpedia.org/ontology/";
String NS_Res = "http://dbpedia.org/resource/";
Modified: trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/TBSLTest.java
===================================================================
--- trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/TBSLTest.java 2013-02-18 14:16:54 UTC (rev 3901)
+++ trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/TBSLTest.java 2013-02-25 11:47:56 UTC (rev 3902)
@@ -3,6 +3,7 @@
import java.io.File;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
+import java.io.FileOutputStream;
import java.net.URL;
import java.util.Collections;
@@ -38,9 +39,9 @@
@Override
protected void setUp() throws Exception {
super.setUp();
- endpoint = new SparqlEndpoint(new URL("http://lgd.aksw.org:8900/sparql"), Collections.singletonList("http://diadem.cs.ox.ac.uk"), Collections.<String>emptyList());
-// model = ModelFactory.createOntologyModel();
-// File dir = new File("/home/lorenz/arbeit/papers/question-answering-iswc-2012/examples/data");
+ endpoint = new SparqlEndpoint(new URL("http://[2001:638:902:2010:0:168:35:138]/sparql"), Collections.singletonList("http://diadem.cs.ox.ac.uk"), Collections.<String>emptyList());
+ model = ModelFactory.createOntologyModel();
+// File dir = new File("/home/me/work/papers/question-answering-iswc-2012/data_v2");
// try {
// for(File f : dir.listFiles()){
// if(f.isFile()){
@@ -53,6 +54,7 @@
// }
// }
// }
+// model.write(new FileOutputStream(dir.getAbsolutePath() + "/oxford-data.ttl"), "TURTLE", null);
// model.read(new FileInputStream(new File("/home/lorenz/arbeit/papers/question-answering-iswc-2012/examples/ontology.ttl")), null, "TURTLE");
// } catch (FileNotFoundException e) {
// e.printStackTrace();
@@ -88,6 +90,7 @@
SPARQLTemplateBasedLearner2 learner = new SPARQLTemplateBasedLearner2(model, resourcesIndex, classesIndex, propertiesIndex);
learner.init();
+ learner.setGrammarFiles(new String[]{"tbsl/lexicon/english.lex","tbsl/lexicon/english_oxford.lex"});
String question = "Give me all houses with more than 3 bathrooms and more than 2 bedrooms.";
@@ -117,9 +120,10 @@
learner.setGrammarFiles(new String[]{"tbsl/lexicon/english.lex","tbsl/lexicon/english_oxford.lex"});
String question = "Give me all houses near a school.";
- question = "Give me all houses with more than 3 bathrooms and more than 2 bedrooms.";
- question = "Give me all Victorian houses in Oxfordshire";
- question = "Edwardian houses close to supermarket for less than 1,000,000 in Oxfordshire";
+ question = "Give me all houses with more than 3 bathrooms.";
+ question = "houses at walking distance from a pharmacy";
+// question = "Give me all Victorian houses in Oxfordshire";
+// question = "Edwardian houses close to supermarket for less than 1,000,000 in Oxfordshire";
// question = "Give me all family houses with more than 2 bathrooms and more than 4 bedrooms";
learner.setQuestion(question);
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-18 14:17:03
|
Revision: 3901
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3901&view=rev
Author: lorenz_b
Date: 2013-02-18 14:16:54 +0000 (Mon, 18 Feb 2013)
Log Message:
-----------
Removed Stanford models because they can be loaded via Maven now.
Removed Paths:
-------------
trunk/components-ext/src/main/resources/tbsl/models/README-Models.txt
trunk/components-ext/src/main/resources/tbsl/models/bidirectional-distsim-wsj-0-18.tagger
trunk/components-ext/src/main/resources/tbsl/models/bidirectional-distsim-wsj-0-18.tagger.props
trunk/components-ext/src/main/resources/tbsl/models/left3words-wsj-0-18.tagger
trunk/components-ext/src/main/resources/tbsl/models/left3words-wsj-0-18.tagger.props
Deleted: trunk/components-ext/src/main/resources/tbsl/models/README-Models.txt
===================================================================
--- trunk/components-ext/src/main/resources/tbsl/models/README-Models.txt 2013-02-18 14:14:10 UTC (rev 3900)
+++ trunk/components-ext/src/main/resources/tbsl/models/README-Models.txt 2013-02-18 14:16:54 UTC (rev 3901)
@@ -1,90 +0,0 @@
-Stanford POS Tagger, v. 3.0.2 - 2011-05-15.
-Copyright (c) 2002-2011 The Board of Trustees of
-The Leland Stanford Junior University. All Rights Reserved.
-
-This document contains (some) information about the models included in
-this release and that may be downloaded for the POS tagger website at
-http://nlp.stanford.edu/software/tagger.shtml . If you have downloaded
-the full tagger, all of the models mentioned in this document are in the
-downloaded package in the same directory as this readme. Otherwise,
-included in the download are two
-English taggers, and the other taggers may be downloaded from the
-website. All taggers are accompanied by the props files used to create
-them; please examine these files for more detailed information about the
-creation of the taggers.
-
-For English, the bidirectional taggers are slightly more accurate, but
-tag much more slowly; choose the appropriate tagger based on your
-speed/performance needs.
-
-English taggers
----------------------------
-bidirectional-distsim-wsj-0-18.tagger
-Trained on WSJ sections 0-18 using a bidirectional architecture and
-including word shape and distributional similarity features.
-Penn Treebank tagset.
-Performance:
-97.28% correct on WSJ 19-21
-(90.46% correct on unknown words)
-
-left3words-wsj-0-18.tagger
-Trained on WSJ sections 0-18 using the left3words architecture and
-includes word shape features. Penn tagset.
-Performance:
-96.97% correct on WSJ 19-21
-(88.85% correct on unknown words)
-
-left3words-distsim-wsj-0-18.tagger
-Trained on WSJ sections 0-18 using the left3words architecture and
-includes word shape and distributional similarity features. Penn tagset.
-Performance:
-97.01% correct on WSJ 19-21
-(89.81% correct on unknown words)
-
-
-Chinese tagger
----------------------------
-chinese.tagger
-Trained on a combination of Chinese Treebank texts from Chinese and Hong
-Kong sources.
-LDC Chinese Treebank POS tag set.
-Performance:
-94.13% on a combination of Chinese and Hong Kong texts
-(78.92% on unknown words)
-
-Arabic tagger
----------------------------
-arabic-accurate.tagger
-Trained on the *entire* ATB p1-3.
-When trained on the train part of the ATB p1-3 split done for the 2005
-JHU Summer Workshop (Diab split), using (augmented) Bies tags, it gets
-the following performance:
-Performance:
-96.50% on dev portion according to Diab split
-(80.59% on unknown words)
-
-arabic-fast.tagger
-4x speed improvement over "accurate".
-Performance:
-96.34% on dev portion according to Diab split
-(80.28% on unknown words)
-
-
-German tagger
----------------------------
-german-accurate.tagger
-Trained on the first 80% of the Negra corpus, which uses the STTS tagset.
-The Stuttgart-Tübingen Tagset (STTS) is a set of 54 tags for annotating
-German text corpora with part-of-speech labels, which was jointly
-developed by the Institut für maschinelle Sprachverarbeitung of the
-University of Stuttgart and the Seminar für Sprachwissenschaft of the
-University of Tübingen. See:
-http://www.ims.uni-stuttgart.de/projekte/CQPDemos/Bundestag/help-tagset.html
-Performance:
-96.90% on the first half of the remaining 20% of the Negra corpus (dev set)
-(90.33% on unknown words)
-
-german-fast.tagger
-8x speed improvement over "accurate".
-Performance:
-96.61% overall / 86.72% unknown.
Deleted: trunk/components-ext/src/main/resources/tbsl/models/bidirectional-distsim-wsj-0-18.tagger
===================================================================
(Binary files differ)
Deleted: trunk/components-ext/src/main/resources/tbsl/models/bidirectional-distsim-wsj-0-18.tagger.props
===================================================================
--- trunk/components-ext/src/main/resources/tbsl/models/bidirectional-distsim-wsj-0-18.tagger.props 2013-02-18 14:14:10 UTC (rev 3900)
+++ trunk/components-ext/src/main/resources/tbsl/models/bidirectional-distsim-wsj-0-18.tagger.props 2013-02-18 14:16:54 UTC (rev 3901)
@@ -1,33 +0,0 @@
-## tagger training invoked at Fri Apr 15 01:00:51 PDT 2011 with arguments:
- model = bidirectional-distsim-wsj-0-18.tagger
- arch = bidirectional5words,naacl2003unknowns,wordshapes(-1,1),distsim(../data/pos-tagger/training/english/egw.bnc.200.pruned,-1,1),distsimconjunction(../data/pos-tagger/training/english/egw.bnc.200.pruned,-1,1)
- trainFile = ../data/pos-tagger/training/english/train-wsj-0-18
- closedClassTags =
- closedClassTagThreshold = 40
- curWordMinFeatureThresh = 2
- debug = false
- debugPrefix =
- tagSeparator = _
- encoding = UTF-8
- iterations = 100
- lang = english
- learnClosedClassTags = false
- minFeatureThresh = 2
- openClassTags =
-rareWordMinFeatureThresh = 5
- rareWordThresh = 5
- search = owlqn
- sgml = false
- sigmaSquared = 0.5
- regL1 = 0.75
- tagInside =
- tokenize = true
- tokenizerFactory =
- tokenizerOptions =
- verbose = false
- verboseResults = true
- veryCommonWordThresh = 250
- xmlInput =
- outputFile =
- outputFormat = slashTags
- outputFormatOptions =
Deleted: trunk/components-ext/src/main/resources/tbsl/models/left3words-wsj-0-18.tagger
===================================================================
(Binary files differ)
Deleted: trunk/components-ext/src/main/resources/tbsl/models/left3words-wsj-0-18.tagger.props
===================================================================
--- trunk/components-ext/src/main/resources/tbsl/models/left3words-wsj-0-18.tagger.props 2013-02-18 14:14:10 UTC (rev 3900)
+++ trunk/components-ext/src/main/resources/tbsl/models/left3words-wsj-0-18.tagger.props 2013-02-18 14:16:54 UTC (rev 3901)
@@ -1,33 +0,0 @@
-## tagger training invoked at Fri Apr 15 07:35:09 PDT 2011 with arguments:
- model = left3words-wsj-0-18.tagger
- arch = left3words,naacl2003unknowns,wordshapes(-1,1)
- trainFile = ../data/pos-tagger/training/english/train-wsj-0-18
- closedClassTags =
- closedClassTagThreshold = 40
- curWordMinFeatureThresh = 2
- debug = false
- debugPrefix =
- tagSeparator = _
- encoding = UTF-8
- iterations = 100
- lang = english
- learnClosedClassTags = false
- minFeatureThresh = 2
- openClassTags =
-rareWordMinFeatureThresh = 10
- rareWordThresh = 5
- search = owlqn
- sgml = false
- sigmaSquared = 0.0
- regL1 = 0.75
- tagInside =
- tokenize = true
- tokenizerFactory =
- tokenizerOptions =
- verbose = false
- verboseResults = true
- veryCommonWordThresh = 250
- xmlInput =
- outputFile =
- outputFormat = slashTags
- outputFormatOptions =
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-18 14:14:18
|
Revision: 3900
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3900&view=rev
Author: lorenz_b
Date: 2013-02-18 14:14:10 +0000 (Mon, 18 Feb 2013)
Log Message:
-----------
Updated index queries.
Modified Paths:
--------------
trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLClassesIndex.java
trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLDatatypePropertiesIndex.java
trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLIndex.java
trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLObjectPropertiesIndex.java
trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoClassesIndex.java
trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoDatatypePropertiesIndex.java
trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoObjectPropertiesIndex.java
trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoPropertiesIndex.java
trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoResourcesIndex.java
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLClassesIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLClassesIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLClassesIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -29,7 +29,7 @@
"FILTER(REGEX(STR(?label), '%s'))}\n" +
"LIMIT %d OFFSET %d";
- super.queryWithLabelTemplate = "SELECT DISTINCT * WHERE {\n" +
+ super.queryWithLabelTemplate = "SELECT DISTINCT ?uri ?label WHERE {\n" +
"?s a ?uri.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label\n" +
"FILTER(REGEX(STR(?label), '%s'))}\n" +
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLDatatypePropertiesIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLDatatypePropertiesIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLDatatypePropertiesIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -23,14 +23,14 @@
private void init(){
super.queryTemplate = "SELECT ?uri WHERE {\n" +
- "?s ?uri ?o.\n" +
+// "?s ?uri ?o.\n" +
"?uri a owl:DatatypeProperty." +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label\n" +
"FILTER(REGEX(STR(?label), '%s', 'i'))}\n" +
"LIMIT %d OFFSET %d";
- super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT * WHERE {\n" +
- "?s ?uri ?o.\n" +
+ super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT ?uri ?label WHERE {\n" +
+// "?s ?uri ?o.\n" +
"?uri a owl:DatatypeProperty." +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label\n" +
"FILTER(REGEX(STR(?label), '%s', 'i'))}\n" +
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -32,7 +32,7 @@
"FILTER(REGEX(STR(?label), '%s'))}\n" +
"LIMIT %d OFFSET %d";
- protected String queryWithLabelTemplate = "SELECT DISTINCT * WHERE {\n" +
+ protected String queryWithLabelTemplate = "SELECT DISTINCT ?uri ?label WHERE {\n" +
"?uri a ?type.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label\n" +
"FILTER(REGEX(STR(?label), '%s'))}\n" +
@@ -139,7 +139,7 @@
}
} else {
rs = QueryExecutionFactory.create(QueryFactory.create(query, Syntax.syntaxARQ), model).execSelect();
- }
+ }System.out.println("Done");
return rs;
}
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLObjectPropertiesIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLObjectPropertiesIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/SPARQLObjectPropertiesIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -23,14 +23,14 @@
private void init(){
super.queryTemplate = "SELECT ?uri WHERE {\n" +
- "?s ?uri ?o.\n" +
+// "?s ?uri ?o.\n" +
"?uri a owl:ObjectProperty." +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label\n" +
"FILTER(REGEX(STR(?label), '%s', 'i'))}\n" +
"LIMIT %d OFFSET %d";
- super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT * WHERE {\n" +
- "?s ?uri ?o.\n" +
+ super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT ?uri ?label WHERE {\n" +
+// "?s ?uri ?o.\n" +
"?uri a owl:ObjectProperty." +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label\n" +
"FILTER(REGEX(STR(?label), '%s', 'i'))}\n" +
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoClassesIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoClassesIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoClassesIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -24,13 +24,15 @@
private void init(){
super.queryTemplate = "SELECT DISTINCT ?uri WHERE {\n" +
- "?s a ?uri.\n" +
+ "{?s a ?uri} UNION {?uri a owl:Class}.\n" +
+// "?s a ?uri.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label.\n" +
"?label bif:contains '\"%s\"'}\n" +
"LIMIT %d OFFSET %d";
- super.queryWithLabelTemplate = "SELECT DISTINCT * WHERE {\n" +
- "?s a ?uri.\n" +
+ super.queryWithLabelTemplate = "SELECT DISTINCT ?uri ?label WHERE {\n" +
+ "{?s a ?uri} UNION {?uri a owl:Class}.\n" +
+// "?s a ?uri.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label.\n" +
"?label bif:contains '\"%s\"'}\n" +
"LIMIT %d OFFSET %d";
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoDatatypePropertiesIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoDatatypePropertiesIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoDatatypePropertiesIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -23,14 +23,14 @@
private void init(){
super.queryTemplate = "SELECT ?uri WHERE {\n" +
- "?s ?uri ?o.\n" +
+// "?s ?uri ?o.\n" +
"?uri a owl:DatatypeProperty.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label.\n" +
"?label bif:contains '\"%s\"'}\n" +
"LIMIT %d OFFSET %d";
- super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT * WHERE {\n" +
- "?s ?uri ?o.\n" +
+ super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT ?uri ?label WHERE {\n" +
+// "?s ?uri ?o.\n" +
"?uri a owl:DatatypeProperty.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label.\n" +
"?label bif:contains '\"%s\"'}\n" +
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoObjectPropertiesIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoObjectPropertiesIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoObjectPropertiesIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -23,14 +23,14 @@
private void init(){
super.queryTemplate = "SELECT ?uri WHERE {\n" +
- "?s ?uri ?o.\n" +
+// "?s ?uri ?o.\n" +
"?uri a owl:ObjectProperty.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label." +
"?label bif:contains '\"%s\"'}\n" +
"LIMIT %d OFFSET %d";
- super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT * WHERE {\n" +
- "?s ?uri ?o.\n" +
+ super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT ?uri ?label WHERE {\n" +
+// "?s ?uri ?o.\n" +
"?uri a owl:ObjectProperty.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label.\n" +
"?label bif:contains '\"%s\"'}\n" +
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoPropertiesIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoPropertiesIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoPropertiesIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -29,13 +29,15 @@
private void init(){
super.queryTemplate = "SELECT DISTINCT ?uri WHERE {\n" +
- "?s ?uri ?o.\n" +
+// "?s ?uri ?o.\n" +
+ "{SELECT DISTINCT ?uri WHERE {?s ?uri ?o.}}\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label.\n" +
"?label bif:contains '\"%s\"'}\n" +
"LIMIT %d OFFSET %d";
super.queryWithLabelTemplate = "PREFIX owl:<http://www.w3.org/2002/07/owl#> SELECT DISTINCT ?uri ?label WHERE {\n" +
- "?s ?uri ?o.\n" +
+// "?s ?uri ?o.\n" +
+ "{SELECT DISTINCT ?uri WHERE {?s ?uri ?o.}}\n" +
// "{?uri a owl:DatatypeProperty.} UNION {?uri a owl:ObjectProperty.} " +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label.\n" +
"?label bif:contains '\"%s\"'}\n" +
Modified: trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoResourcesIndex.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoResourcesIndex.java 2013-02-16 13:03:00 UTC (rev 3899)
+++ trunk/components-ext/src/main/java/org/dllearner/common/index/VirtuosoResourcesIndex.java 2013-02-18 14:14:10 UTC (rev 3900)
@@ -32,7 +32,7 @@
"?label bif:contains '\"%s\"'}\n" +
"LIMIT %d OFFSET %d";
- super.queryWithLabelTemplate = "SELECT DISTINCT * WHERE {\n" +
+ super.queryWithLabelTemplate = "SELECT DISTINCT ?uri ?label WHERE {\n" +
"?uri a ?type.\n" +
"?uri <http://www.w3.org/2000/01/rdf-schema#label> ?label.\n" +
"?label bif:contains '\"%s\"'}\n" +
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-16 13:03:08
|
Revision: 3899
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3899&view=rev
Author: lorenz_b
Date: 2013-02-16 13:03:00 +0000 (Sat, 16 Feb 2013)
Log Message:
-----------
Added option to work on JENA model.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/cli/CLI.java
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner2.java
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner3.java
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/Knowledgebase.java
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/PopularityMap.java
trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/Evaluation.java
trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner3Test.java
Added Paths:
-----------
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/LocalKnowledgebase.java
trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/RemoteKnowledgebase.java
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java 2013-02-11 12:18:04 UTC (rev 3898)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -68,7 +68,9 @@
import org.dllearner.utilities.Helper;
import org.springframework.beans.factory.annotation.Autowired;
+import com.hp.hpl.jena.query.QueryExecutionFactory;
import com.hp.hpl.jena.query.QuerySolution;
+import com.hp.hpl.jena.query.ResultSet;
import com.hp.hpl.jena.query.ResultSetRewindable;
import com.hp.hpl.jena.rdf.model.Model;
import com.hp.hpl.jena.rdf.model.Statement;
@@ -94,6 +96,7 @@
// private QTLConfigurator configurator;
private SparqlEndpoint endpoint;
+ private Model model;
private ExtractionDBCache cache;
private QueryTreeCache treeCache;
@@ -157,6 +160,21 @@
negExampleTrees = new ArrayList<QueryTree<String>>();
}
+ public QTL(Model model) {
+ this.model = model;
+
+ treeCache = new QueryTreeCache();
+ cbdGenerator = new CachingConciseBoundedDescriptionGenerator(new ConciseBoundedDescriptionGeneratorImpl(model));
+ cbdGenerator.setRecursionDepth(maxQueryTreeDepth);
+
+ lggGenerator = new LGGGeneratorImpl<String>();
+ nbr = new NBR<String>(model);
+ nbr.setMaxExecutionTimeInSeconds(maxExecutionTimeInSeconds);
+
+ posExampleTrees = new ArrayList<QueryTree<String>>();
+ negExampleTrees = new ArrayList<QueryTree<String>>();
+ }
+
public String getQuestion(List<String> posExamples, List<String> negExamples) throws EmptyLGGException, NegativeTreeCoverageExecption, TimeOutException {
this.posExamples = posExamples;
this.negExamples = negExamples;
@@ -312,8 +330,13 @@
private SortedSet<String> getResources(QueryTree<String> tree){
SortedSet<String> resources = new TreeSet<String>();
String query = getDistinctSPARQLQuery(tree);
- String result = cache.executeSelectQuery(endpoint, query);
- ResultSetRewindable rs = SparqlQuery.convertJSONtoResultSet(result);
+ ResultSet rs;
+ if(endpoint != null){
+ rs = SparqlQuery.convertJSONtoResultSet(cache.executeSelectQuery(endpoint, query));
+ } else {
+ rs = QueryExecutionFactory.create(query, model).execSelect();
+ }
+
String uri;
QuerySolution qs;
while(rs.hasNext()){
Modified: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/cli/CLI.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/cli/CLI.java 2013-02-11 12:18:04 UTC (rev 3898)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/cli/CLI.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -10,6 +10,7 @@
import org.dllearner.algorithm.tbsl.learning.NoTemplateFoundException;
import org.dllearner.algorithm.tbsl.learning.SPARQLTemplateBasedLearner2;
import org.dllearner.algorithm.tbsl.util.Knowledgebase;
+import org.dllearner.algorithm.tbsl.util.RemoteKnowledgebase;
import org.dllearner.common.index.Index;
import org.dllearner.common.index.SOLRIndex;
import org.dllearner.kb.sparql.SparqlEndpoint;
@@ -30,7 +31,7 @@
Index propertiesIndex = new SOLRIndex("http://dbpedia.aksw.org:8080/solr/dbpedia_properties");
- Knowledgebase kb = new Knowledgebase(endpoint, "DBpedia Live", "TODO", resourcesIndex, propertiesIndex, classesIndex, null);
+ Knowledgebase kb = new RemoteKnowledgebase(endpoint, "DBpedia Live", "TODO", resourcesIndex, propertiesIndex, classesIndex, null);
SPARQLTemplateBasedLearner2 learner = new SPARQLTemplateBasedLearner2(kb);
System.out.println("======= TBSL v0.1 =============");
Modified: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner2.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner2.java 2013-02-11 12:18:04 UTC (rev 3898)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner2.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -44,7 +44,9 @@
import org.dllearner.algorithm.tbsl.sparql.WeightedQuery;
import org.dllearner.algorithm.tbsl.templator.Templator;
import org.dllearner.algorithm.tbsl.util.Knowledgebase;
+import org.dllearner.algorithm.tbsl.util.LocalKnowledgebase;
import org.dllearner.algorithm.tbsl.util.PopularityMap;
+import org.dllearner.algorithm.tbsl.util.RemoteKnowledgebase;
import org.dllearner.algorithm.tbsl.util.PopularityMap.EntityType;
import org.dllearner.algorithm.tbsl.util.Similarity;
import org.dllearner.algorithm.tbsl.util.UnknownPropertyHelper.SymPropertyDirection;
@@ -163,12 +165,44 @@
}
public SPARQLTemplateBasedLearner2(Knowledgebase knowledgebase, PartOfSpeechTagger posTagger, WordNet wordNet, Options options){
- this(knowledgebase.getEndpoint(), knowledgebase.getResourceIndex(), knowledgebase.getClassIndex(),knowledgebase.getPropertyIndex(), posTagger, wordNet, options);
+ this(knowledgebase, posTagger, wordNet, options, null);
}
public SPARQLTemplateBasedLearner2(Knowledgebase knowledgebase){
- this(knowledgebase.getEndpoint(), knowledgebase.getResourceIndex(), knowledgebase.getClassIndex(),knowledgebase.getPropertyIndex(), new StanfordPartOfSpeechTagger(), new WordNet(), new Options());
+ this(knowledgebase, new StanfordPartOfSpeechTagger(), new WordNet(), new Options());
}
+
+ public SPARQLTemplateBasedLearner2(Knowledgebase knowledgebase, PartOfSpeechTagger posTagger, WordNet wordNet, Options options, ExtractionDBCache cache){
+ if(knowledgebase instanceof LocalKnowledgebase){
+ this.model = ((LocalKnowledgebase) knowledgebase).getModel();
+ } else {
+ this.endpoint = ((RemoteKnowledgebase) knowledgebase).getEndpoint();
+ }
+ this.resourcesIndex = knowledgebase.getResourceIndex();
+ this.classesIndex = knowledgebase.getClassIndex();
+ this.propertiesIndex = knowledgebase.getPropertyIndex();
+ this.mappingIndex = knowledgebase.getMappingIndex();
+ if(propertiesIndex instanceof SPARQLPropertiesIndex){
+ if(propertiesIndex instanceof VirtuosoPropertiesIndex){
+ datatypePropertiesIndex = new VirtuosoDatatypePropertiesIndex((SPARQLPropertiesIndex)propertiesIndex);
+ objectPropertiesIndex = new VirtuosoObjectPropertiesIndex((SPARQLPropertiesIndex)propertiesIndex);
+ } else {
+ datatypePropertiesIndex = new SPARQLDatatypePropertiesIndex((SPARQLPropertiesIndex)propertiesIndex);
+ objectPropertiesIndex = new SPARQLObjectPropertiesIndex((SPARQLPropertiesIndex)propertiesIndex);
+ }
+ } else {
+ datatypePropertiesIndex = propertiesIndex;
+ objectPropertiesIndex = propertiesIndex;
+ }
+ this.posTagger = posTagger;
+ this.wordNet = wordNet;
+ this.cache = cache;
+
+ reasoner = new SPARQLReasoner(new SparqlEndpointKS(endpoint));
+ setOptions(options);
+
+ setMappingIndex(knowledgebase.getMappingIndex());
+ }
public SPARQLTemplateBasedLearner2(SparqlEndpoint endpoint, Index index){
this(endpoint, index, new StanfordPartOfSpeechTagger());
@@ -301,7 +335,11 @@
}
public void setKnowledgebase(Knowledgebase knowledgebase){
- this.endpoint = knowledgebase.getEndpoint();
+ if(knowledgebase instanceof LocalKnowledgebase){
+ this.model = ((LocalKnowledgebase) knowledgebase).getModel();
+ } else {
+ this.endpoint = ((RemoteKnowledgebase) knowledgebase).getEndpoint();
+ }
this.resourcesIndex = knowledgebase.getResourceIndex();
this.classesIndex = knowledgebase.getClassIndex();
this.propertiesIndex = knowledgebase.getPropertyIndex();
Modified: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner3.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner3.java 2013-02-11 12:18:04 UTC (rev 3898)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner3.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -46,7 +46,9 @@
import org.dllearner.algorithm.tbsl.sparql.WeightedQuery;
import org.dllearner.algorithm.tbsl.templator.Templator;
import org.dllearner.algorithm.tbsl.util.Knowledgebase;
+import org.dllearner.algorithm.tbsl.util.LocalKnowledgebase;
import org.dllearner.algorithm.tbsl.util.PopularityMap;
+import org.dllearner.algorithm.tbsl.util.RemoteKnowledgebase;
import org.dllearner.algorithm.tbsl.util.PopularityMap.EntityType;
import org.dllearner.algorithm.tbsl.util.SPARQLEndpointMetrics;
import org.dllearner.algorithm.tbsl.util.Similarity;
@@ -171,17 +173,44 @@
}
public SPARQLTemplateBasedLearner3(Knowledgebase knowledgebase, PartOfSpeechTagger posTagger, WordNet wordNet, Options options){
- this(knowledgebase.getEndpoint(), knowledgebase.getResourceIndex(), knowledgebase.getClassIndex(),knowledgebase.getPropertyIndex(), posTagger, wordNet, options);
+ this(knowledgebase, posTagger, wordNet, options, null);
setMappingIndex(knowledgebase.getMappingIndex());
}
public SPARQLTemplateBasedLearner3(Knowledgebase knowledgebase, PartOfSpeechTagger posTagger, WordNet wordNet, Options options, ExtractionDBCache cache){
- this(knowledgebase.getEndpoint(), knowledgebase.getResourceIndex(), knowledgebase.getClassIndex(),knowledgebase.getPropertyIndex(), posTagger, wordNet, options, cache);
+ if(knowledgebase instanceof LocalKnowledgebase){
+ this.model = ((LocalKnowledgebase) knowledgebase).getModel();
+ } else {
+ this.endpoint = ((RemoteKnowledgebase) knowledgebase).getEndpoint();
+ }
+ this.resourcesIndex = knowledgebase.getResourceIndex();
+ this.classesIndex = knowledgebase.getClassIndex();
+ this.propertiesIndex = knowledgebase.getPropertyIndex();
+ this.mappingIndex = knowledgebase.getMappingIndex();
+ if(propertiesIndex instanceof SPARQLPropertiesIndex){
+ if(propertiesIndex instanceof VirtuosoPropertiesIndex){
+ datatypePropertiesIndex = new VirtuosoDatatypePropertiesIndex((SPARQLPropertiesIndex)propertiesIndex);
+ objectPropertiesIndex = new VirtuosoObjectPropertiesIndex((SPARQLPropertiesIndex)propertiesIndex);
+ } else {
+ datatypePropertiesIndex = new SPARQLDatatypePropertiesIndex((SPARQLPropertiesIndex)propertiesIndex);
+ objectPropertiesIndex = new SPARQLObjectPropertiesIndex((SPARQLPropertiesIndex)propertiesIndex);
+ }
+ } else {
+ datatypePropertiesIndex = propertiesIndex;
+ objectPropertiesIndex = propertiesIndex;
+ }
+ this.posTagger = posTagger;
+ this.wordNet = wordNet;
+ this.cache = cache;
+
+ reasoner = new SPARQLReasoner(new SparqlEndpointKS(endpoint));
+ setOptions(options);
+
setMappingIndex(knowledgebase.getMappingIndex());
}
public SPARQLTemplateBasedLearner3(Knowledgebase knowledgebase){
- this(knowledgebase.getEndpoint(), knowledgebase.getResourceIndex(), knowledgebase.getClassIndex(),knowledgebase.getPropertyIndex(), new StanfordPartOfSpeechTagger(), new WordNet(), new Options());
+ this(knowledgebase, new StanfordPartOfSpeechTagger(), new WordNet(), new Options());
setMappingIndex(knowledgebase.getMappingIndex());
}
@@ -303,7 +332,11 @@
}
public void setKnowledgebase(Knowledgebase knowledgebase){
- this.endpoint = knowledgebase.getEndpoint();
+ if(knowledgebase instanceof LocalKnowledgebase){
+ this.model = ((LocalKnowledgebase) knowledgebase).getModel();
+ } else {
+ this.endpoint = ((RemoteKnowledgebase) knowledgebase).getEndpoint();
+ }
this.resourcesIndex = knowledgebase.getResourceIndex();
this.classesIndex = knowledgebase.getClassIndex();
this.propertiesIndex = knowledgebase.getPropertyIndex();
Modified: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/Knowledgebase.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/Knowledgebase.java 2013-02-11 12:18:04 UTC (rev 3898)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/Knowledgebase.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -2,12 +2,10 @@
import org.dllearner.common.index.Index;
import org.dllearner.common.index.MappingBasedIndex;
-import org.dllearner.kb.sparql.SparqlEndpoint;
-public class Knowledgebase {
+public abstract class Knowledgebase {
private String label;
- private SparqlEndpoint endpoint;
private String description;
private Index resourceIndex;
@@ -16,30 +14,25 @@
private MappingBasedIndex mappingIndex;
- public Knowledgebase(SparqlEndpoint endpoint, String label, String description,
+ public Knowledgebase(String label, String description,
Index resourceIndex, Index propertyIndex, Index classIndex) {
- this(endpoint, label, description, resourceIndex, propertyIndex, classIndex, null);
+ this(label, description, resourceIndex, propertyIndex, classIndex, null);
}
- public Knowledgebase(SparqlEndpoint endpoint, String label, String description,
+ public Knowledgebase(String label, String description,
Index resourceIndex, Index propertyIndex, Index classIndex, MappingBasedIndex mappingIndex) {
this.label = label;
- this.endpoint = endpoint;
this.description = description;
this.resourceIndex = resourceIndex;
this.propertyIndex = propertyIndex;
this.classIndex = classIndex;
this.mappingIndex = mappingIndex;
}
-
+
public String getLabel() {
return label;
}
- public SparqlEndpoint getEndpoint() {
- return endpoint;
- }
-
public String getDescription() {
return description;
}
Added: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/LocalKnowledgebase.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/LocalKnowledgebase.java (rev 0)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/LocalKnowledgebase.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -0,0 +1,22 @@
+package org.dllearner.algorithm.tbsl.util;
+
+import org.dllearner.common.index.Index;
+import org.dllearner.common.index.MappingBasedIndex;
+
+import com.hp.hpl.jena.rdf.model.Model;
+
+public class LocalKnowledgebase extends Knowledgebase{
+
+ private Model model;
+
+ public LocalKnowledgebase(Model model, String label, String description, Index resourceIndex, Index propertyIndex,
+ Index classIndex, MappingBasedIndex mappingIndex) {
+ super(label, description, resourceIndex, propertyIndex, classIndex, mappingIndex);
+ this.model = model;
+ }
+
+ public Model getModel() {
+ return model;
+ }
+
+}
Modified: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/PopularityMap.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/PopularityMap.java 2013-02-11 12:18:04 UTC (rev 3898)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/PopularityMap.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -20,8 +20,10 @@
import org.dllearner.kb.sparql.SparqlEndpoint;
import org.dllearner.kb.sparql.SparqlQuery;
+import com.hp.hpl.jena.query.QueryExecutionFactory;
import com.hp.hpl.jena.query.QuerySolution;
import com.hp.hpl.jena.query.ResultSet;
+import com.hp.hpl.jena.rdf.model.Model;
public class PopularityMap {
@@ -30,6 +32,7 @@
}
private SparqlEndpoint endpoint;
+ private Model model;
private ExtractionDBCache cache;
private String file;
@@ -41,9 +44,13 @@
this.file = file;
this.endpoint = endpoint;
this.cache = cache;
-
}
+ public PopularityMap(String file, Model model) {
+ this.file = file;
+ this.model = model;
+ }
+
public void init() {
boolean deserialized = deserialize();
if(!deserialized){
@@ -137,7 +144,12 @@
query = String.format("SELECT COUNT(*) WHERE {?s ?p <%s>}", uri);
}
int pop = 0;
- ResultSet rs = SparqlQuery.convertJSONtoResultSet(cache.executeSelectQuery(endpoint, query));
+ ResultSet rs;
+ if(endpoint != null){
+ rs = SparqlQuery.convertJSONtoResultSet(cache.executeSelectQuery(endpoint, query));
+ } else {
+ rs = QueryExecutionFactory.create(query, model).execSelect();
+ }
QuerySolution qs;
String projectionVar;
while(rs.hasNext()){
Added: trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/RemoteKnowledgebase.java
===================================================================
--- trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/RemoteKnowledgebase.java (rev 0)
+++ trunk/components-ext/src/main/java/org/dllearner/algorithm/tbsl/util/RemoteKnowledgebase.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -0,0 +1,22 @@
+package org.dllearner.algorithm.tbsl.util;
+
+import org.dllearner.common.index.Index;
+import org.dllearner.common.index.MappingBasedIndex;
+import org.dllearner.kb.sparql.SparqlEndpoint;
+
+public class RemoteKnowledgebase extends Knowledgebase{
+
+ private SparqlEndpoint endpoint;
+
+ public RemoteKnowledgebase(SparqlEndpoint endpoint, String label, String description, Index resourceIndex, Index propertyIndex,
+ Index classIndex, MappingBasedIndex mappingIndex) {
+ super(label, description, resourceIndex, propertyIndex, classIndex, mappingIndex);
+ this.endpoint = endpoint;
+ }
+
+ public SparqlEndpoint getEndpoint() {
+ return endpoint;
+ }
+
+
+}
Modified: trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/Evaluation.java
===================================================================
--- trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/Evaluation.java 2013-02-11 12:18:04 UTC (rev 3898)
+++ trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/Evaluation.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -42,6 +42,7 @@
import org.dllearner.algorithm.tbsl.sparql.WeightedQuery;
import org.dllearner.algorithm.tbsl.util.Knowledgebase;
import org.dllearner.algorithm.tbsl.util.LatexWriter;
+import org.dllearner.algorithm.tbsl.util.RemoteKnowledgebase;
import org.dllearner.common.index.Index;
import org.dllearner.common.index.SOLRIndex;
import org.dllearner.kb.sparql.ExtractionDBCache;
@@ -99,7 +100,7 @@
Index propertiesIndex = new SOLRIndex("http://dbpedia.aksw.org:8080/solr/dbpedia_properties");
- Knowledgebase kb = new Knowledgebase(endpoint, "DBpedia Live", "TODO", resourcesIndex, propertiesIndex, classesIndex, null);
+ Knowledgebase kb = new RemoteKnowledgebase(endpoint, "DBpedia Live", "TODO", resourcesIndex, propertiesIndex, classesIndex, null);
stbl = new SPARQLTemplateBasedLearner2(kb);
stbl.setUseIdealTagger(USE_IDEAL_TAGGER);
Modified: trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner3Test.java
===================================================================
--- trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner3Test.java 2013-02-11 12:18:04 UTC (rev 3898)
+++ trunk/components-ext/src/test/java/org/dllearner/algorithm/tbsl/learning/SPARQLTemplateBasedLearner3Test.java 2013-02-16 13:03:00 UTC (rev 3899)
@@ -55,6 +55,7 @@
import org.dllearner.algorithm.tbsl.nlp.WordNet;
import org.dllearner.algorithm.tbsl.templator.Templator;
import org.dllearner.algorithm.tbsl.util.Knowledgebase;
+import org.dllearner.algorithm.tbsl.util.RemoteKnowledgebase;
import org.dllearner.common.index.HierarchicalIndex;
import org.dllearner.common.index.Index;
import org.dllearner.common.index.MappingBasedIndex;
@@ -794,7 +795,7 @@
SPARQLTemplateBasedLearner2.class.getClassLoader().getResource("test/dbpedia_objectproperty_mappings.txt").getPath()
);
- Knowledgebase kb = new Knowledgebase(dbpediaLiveEndpoint, "DBpedia Live", "TODO", resourcesIndex, propertiesIndex, classesIndex, mappingIndex);
+ Knowledgebase kb = new RemoteKnowledgebase(dbpediaLiveEndpoint, "DBpedia Live", "TODO", resourcesIndex, propertiesIndex, classesIndex, mappingIndex);
return kb;
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-11 12:18:12
|
Revision: 3898
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3898&view=rev
Author: lorenz_b
Date: 2013-02-11 12:18:04 +0000 (Mon, 11 Feb 2013)
Log Message:
-----------
Changed datatype URI to IRI to avoid problems with special chars.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/core/owl/Datatype.java
trunk/components-core/src/main/java/org/dllearner/core/owl/OWL2Datatype.java
trunk/components-core/src/main/java/org/dllearner/core/owl/TypedConstant.java
trunk/components-core/src/main/java/org/dllearner/reasoning/OWLAPIReasoner.java
trunk/components-core/src/main/java/org/dllearner/reasoning/PelletReasoner.java
trunk/components-core/src/main/java/org/dllearner/reasoning/fuzzydll/FuzzyOWLAPIReasoner.java
trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIAxiomConvertVisitor.java
trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIConverter.java
trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIDescriptionConvertVisitor.java
Modified: trunk/components-core/src/main/java/org/dllearner/core/owl/Datatype.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/core/owl/Datatype.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/core/owl/Datatype.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -19,7 +19,6 @@
package org.dllearner.core.owl;
-import java.net.URI;
import java.util.Map;
import org.semanticweb.owlapi.model.IRI;
@@ -31,19 +30,19 @@
*/
public class Datatype implements DataRange, Comparable<Datatype> {
- private URI uri;
+ private IRI iri;
- public Datatype(String uriString) {
- uri = URI.create(uriString);
+ public Datatype(String iriString) {
+ iri = IRI.create(iriString);
}
- public URI getURI() {
- return uri;
+ public IRI getIRI() {
+ return iri;
}
@Override
public String toString() {
- return uri.toString();
+ return iri.toString();
}
public int getLength() {
@@ -55,11 +54,11 @@
}
public String toString(String baseURI, Map<String, String> prefixes) {
- return uri.toString();
+ return iri.toString();
}
public String toKBSyntaxString(String baseURI, Map<String, String> prefixes) {
- return uri.toString();
+ return iri.toString();
}
/* (non-Javadoc)
@@ -67,14 +66,14 @@
*/
@Override
public String toManchesterSyntaxString(String baseURI, Map<String, String> prefixes) {
- return uri.toString();
+ return iri.toString();
}
@Override
public int hashCode() {
final int prime = 31;
int result = 1;
- result = prime * result + ((uri == null) ? 0 : uri.hashCode());
+ result = prime * result + ((iri == null) ? 0 : iri.hashCode());
return result;
}
@@ -87,16 +86,16 @@
if (getClass() != obj.getClass())
return false;
Datatype other = (Datatype) obj;
- if (uri == null) {
- if (other.uri != null)
+ if (iri == null) {
+ if (other.iri != null)
return false;
- } else if (!uri.equals(other.uri))
+ } else if (!iri.equals(other.iri))
return false;
return true;
}
@Override
public int compareTo(Datatype o) {
- return this.getURI().compareTo(o.getURI());
+ return this.getIRI().compareTo(o.getIRI());
}
}
Modified: trunk/components-core/src/main/java/org/dllearner/core/owl/OWL2Datatype.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/core/owl/OWL2Datatype.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/core/owl/OWL2Datatype.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -19,7 +19,7 @@
package org.dllearner.core.owl;
-import java.net.URI;
+import org.semanticweb.owlapi.model.IRI;
public enum OWL2Datatype {
@@ -41,8 +41,8 @@
return datatype;
}
- public URI getURI() {
- return datatype.getURI();
+ public IRI getIRI() {
+ return datatype.getIRI();
}
}
Modified: trunk/components-core/src/main/java/org/dllearner/core/owl/TypedConstant.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/core/owl/TypedConstant.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/core/owl/TypedConstant.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -99,7 +99,7 @@
*/
public int compareTo(TypedConstant o) {
// the first criteria is the datatype
- int datatypeComparision = datatype.getURI().compareTo(datatype.getURI());
+ int datatypeComparision = datatype.getIRI().compareTo(datatype.getIRI());
if(datatypeComparision == 0) {
// the second criterion is the literal value
return literal.compareTo(o.literal);
Modified: trunk/components-core/src/main/java/org/dllearner/reasoning/OWLAPIReasoner.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/reasoning/OWLAPIReasoner.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/reasoning/OWLAPIReasoner.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -20,7 +20,6 @@
package org.dllearner.reasoning;
import java.io.File;
-import java.net.URI;
import java.net.URL;
import java.util.ArrayList;
import java.util.Arrays;
@@ -379,19 +378,23 @@
for (OWLDataProperty owlProperty : owlDatatypeProperties) {
DatatypeProperty dtp = new DatatypeProperty(owlProperty.toStringID());
Set<OWLDataRange> ranges = owlProperty.getRanges(allImports);
- Iterator<OWLDataRange> it = ranges.iterator();
- if (it.hasNext()) {
- OWLDataRange range = it.next();
- if (range.isDatatype()) {
- URI uri = ((OWLDatatype) range).getIRI().toURI();
- if (uri.equals(OWL2Datatype.BOOLEAN.getURI())) booleanDatatypeProperties.add(dtp);
- else if (uri.equals(OWL2Datatype.DOUBLE.getURI())) doubleDatatypeProperties.add(dtp);
- else if (uri.equals(OWL2Datatype.INT.getURI())) intDatatypeProperties.add(dtp);
- else if (uri.equals(OWL2Datatype.STRING.getURI())) stringDatatypeProperties.add(dtp);
- }
- } else {
- stringDatatypeProperties.add(dtp);
- }
+ Iterator<OWLDataRange> it = ranges.iterator();
+ if (it.hasNext()) {
+ OWLDataRange range = it.next();
+ if (range.isDatatype()) {
+ IRI iri = ((OWLDatatype) range).getIRI();
+ if (iri.equals(OWL2Datatype.BOOLEAN.getIRI()))
+ booleanDatatypeProperties.add(dtp);
+ else if (iri.equals(OWL2Datatype.DOUBLE.getIRI()))
+ doubleDatatypeProperties.add(dtp);
+ else if (iri.equals(OWL2Datatype.INT.getIRI()))
+ intDatatypeProperties.add(dtp);
+ else if (iri.equals(OWL2Datatype.STRING.getIRI()))
+ stringDatatypeProperties.add(dtp);
+ }
+ } else {
+ stringDatatypeProperties.add(dtp);
+ }
datatypeProperties.add(dtp);
}
for (OWLNamedIndividual owlIndividual : owlIndividuals) {
Modified: trunk/components-core/src/main/java/org/dllearner/reasoning/PelletReasoner.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/reasoning/PelletReasoner.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/reasoning/PelletReasoner.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -20,7 +20,6 @@
package org.dllearner.reasoning;
import java.io.File;
-import java.net.URI;
import java.net.URISyntaxException;
import java.util.Collection;
import java.util.Collections;
@@ -255,14 +254,14 @@
if(it.hasNext()) {
OWLDataRange range = it.next();
if(range.isDatatype()) {
- URI uri = ((OWLDatatype)range).getIRI().toURI();
- if(uri.equals(OWL2Datatype.BOOLEAN.getURI()))
+ IRI iri = ((OWLDatatype)range).getIRI();
+ if(iri.equals(OWL2Datatype.BOOLEAN.getIRI()))
booleanDatatypeProperties.add(dtp);
- else if(uri.equals(OWL2Datatype.DOUBLE.getURI()))
+ else if(iri.equals(OWL2Datatype.DOUBLE.getIRI()))
doubleDatatypeProperties.add(dtp);
- else if(uri.equals(OWL2Datatype.INT.getURI()))
+ else if(iri.equals(OWL2Datatype.INT.getIRI()))
intDatatypeProperties.add(dtp);
- else if(uri.equals(OWL2Datatype.STRING.getURI()))
+ else if(iri.equals(OWL2Datatype.STRING.getIRI()))
stringDatatypeProperties.add(dtp);
}
}
Modified: trunk/components-core/src/main/java/org/dllearner/reasoning/fuzzydll/FuzzyOWLAPIReasoner.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/reasoning/fuzzydll/FuzzyOWLAPIReasoner.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/reasoning/fuzzydll/FuzzyOWLAPIReasoner.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -21,7 +21,6 @@
import java.io.File;
import java.net.MalformedURLException;
-import java.net.URI;
import java.net.URL;
import java.util.Collection;
import java.util.Collections;
@@ -116,8 +115,6 @@
import org.semanticweb.owlapi.util.SimpleIRIMapper;
import org.semanticweb.owlapi.vocab.PrefixOWLOntologyFormat;
-import uk.ac.manchester.cs.factplusplus.owlapiv3.FaCTPlusPlusReasonerFactory;
-
import com.clarkparsia.pellet.owlapiv3.PelletReasonerFactory;
/**
@@ -385,14 +382,14 @@
if(it.hasNext()) {
OWLDataRange range = it.next();
if(range.isDatatype()) {
- URI uri = ((OWLDatatype)range).getIRI().toURI();
- if(uri.equals(OWL2Datatype.BOOLEAN.getURI()))
+ IRI iri = ((OWLDatatype)range).getIRI();
+ if(iri.equals(OWL2Datatype.BOOLEAN.getIRI()))
booleanDatatypeProperties.add(dtp);
- else if(uri.equals(OWL2Datatype.DOUBLE.getURI()))
+ else if(iri.equals(OWL2Datatype.DOUBLE.getIRI()))
doubleDatatypeProperties.add(dtp);
- else if(uri.equals(OWL2Datatype.INT.getURI()))
+ else if(iri.equals(OWL2Datatype.INT.getIRI()))
intDatatypeProperties.add(dtp);
- else if(uri.equals(OWL2Datatype.STRING.getURI()))
+ else if(iri.equals(OWL2Datatype.STRING.getIRI()))
stringDatatypeProperties.add(dtp);
}
} else {
Modified: trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIAxiomConvertVisitor.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIAxiomConvertVisitor.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIAxiomConvertVisitor.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -346,7 +346,7 @@
public void visit(DatatypePropertyRangeAxiom axiom) {
DataRange dr = axiom.getRange();
Datatype dt = (Datatype) dr;
- OWLDatatype odt = factory.getOWLDatatype(IRI.create(dt.getURI()));
+ OWLDatatype odt = factory.getOWLDatatype(dt.getIRI());
OWLDataProperty dp = factory.getOWLDataProperty(IRI.create(axiom.getProperty().getName()));
OWLAxiom axiomOWLAPI = factory.getOWLDataPropertyRangeAxiom(dp, odt);
addAxiom(axiomOWLAPI);
Modified: trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIConverter.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIConverter.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIConverter.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -19,7 +19,6 @@
package org.dllearner.utilities.owl;
-import java.net.URI;
import java.util.Set;
import java.util.SortedSet;
import java.util.TreeSet;
@@ -177,20 +176,20 @@
}
public static Datatype convertDatatype(OWLDatatype dataType) {
- URI uri = dataType.getIRI().toURI();
- if(uri.equals(OWL2Datatype.BOOLEAN.getURI()))
+ IRI iri = dataType.getIRI();
+ if(iri.equals(OWL2Datatype.BOOLEAN.getIRI()))
return OWL2Datatype.BOOLEAN.getDatatype();
- else if(uri.equals(OWL2Datatype.DOUBLE.getURI()))
+ else if(iri.equals(OWL2Datatype.DOUBLE.getIRI()))
return OWL2Datatype.DOUBLE.getDatatype();
- else if(uri.equals(OWL2Datatype.INT.getURI()))
+ else if(iri.equals(OWL2Datatype.INT.getIRI()))
return OWL2Datatype.INT.getDatatype();
- else if(uri.equals(OWL2Datatype.INTEGER.getURI()))
+ else if(iri.equals(OWL2Datatype.INTEGER.getIRI()))
return OWL2Datatype.INTEGER.getDatatype();
- else if(uri.equals(OWL2Datatype.STRING.getURI()))
+ else if(iri.equals(OWL2Datatype.STRING.getIRI()))
return OWL2Datatype.STRING.getDatatype();
- else if(uri.equals(OWL2Datatype.DATE.getURI()))
+ else if(iri.equals(OWL2Datatype.DATE.getIRI()))
return OWL2Datatype.DATE.getDatatype();
- else if(uri.equals(OWL2Datatype.DATETIME.getURI()))
+ else if(iri.equals(OWL2Datatype.DATETIME.getIRI()))
return OWL2Datatype.DATETIME.getDatatype();
throw new Error("Unsupported datatype " + dataType + ". Please inform a DL-Learner developer to add it.");
}
Modified: trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIDescriptionConvertVisitor.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIDescriptionConvertVisitor.java 2013-02-07 14:03:54 UTC (rev 3897)
+++ trunk/components-core/src/main/java/org/dllearner/utilities/owl/OWLAPIDescriptionConvertVisitor.java 2013-02-11 12:18:04 UTC (rev 3898)
@@ -340,7 +340,7 @@
else if(datatype.equals(OWL2Datatype.DOUBLE.getDatatype()))
return factory.getDoubleOWLDatatype();
else if(datatype.equals(OWL2Datatype.STRING.getDatatype()))
- return factory.getOWLDatatype(IRI.create(OWL2Datatype.STRING.getURI()));
+ return factory.getOWLDatatype(OWL2Datatype.STRING.getIRI());
throw new Error("OWLAPIDescriptionConverter: datatype "+datatype+" not implemented");
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-07 14:04:07
|
Revision: 3897
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3897&view=rev
Author: lorenz_b
Date: 2013-02-07 14:03:54 +0000 (Thu, 07 Feb 2013)
Log Message:
-----------
Cont. matching.
Modified Paths:
--------------
trunk/scripts/pom.xml
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java
Modified: trunk/scripts/pom.xml
===================================================================
--- trunk/scripts/pom.xml 2013-02-06 18:17:28 UTC (rev 3896)
+++ trunk/scripts/pom.xml 2013-02-07 14:03:54 UTC (rev 3897)
@@ -116,6 +116,7 @@
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
<artifactId>owlapi-distribution</artifactId>
+ <version>3.4.4-SNAPSHOT</version>
</dependency>
<dependency>
<groupId>net.sourceforge.owlapi</groupId>
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-02-06 18:17:28 UTC (rev 3896)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-02-07 14:03:54 UTC (rev 3897)
@@ -19,15 +19,18 @@
import org.apache.log4j.Logger;
import org.dllearner.algorithms.celoe.CELOE;
+import org.dllearner.algorithms.qtl.QTL;
import org.dllearner.core.AbstractLearningProblem;
import org.dllearner.core.AbstractReasonerComponent;
import org.dllearner.core.ComponentInitException;
import org.dllearner.core.EvaluatedDescription;
import org.dllearner.core.KnowledgeSource;
+import org.dllearner.core.LearningProblemUnsupportedException;
import org.dllearner.core.owl.Description;
import org.dllearner.core.owl.Individual;
import org.dllearner.core.owl.NamedClass;
import org.dllearner.core.owl.ObjectProperty;
+import org.dllearner.kb.LocalModelBasedSparqlEndpointKS;
import org.dllearner.kb.OWLAPIOntology;
import org.dllearner.kb.SparqlEndpointKS;
import org.dllearner.kb.sparql.ConciseBoundedDescriptionGenerator;
@@ -48,15 +51,16 @@
import org.dllearner.utilities.examples.AutomaticNegativeExampleFinderSPARQL2;
import org.dllearner.utilities.owl.OWLAPIDescriptionConvertVisitor;
import org.semanticweb.owlapi.apibinding.OWLManager;
-import org.semanticweb.owlapi.io.UnparsableOntologyException;
import org.semanticweb.owlapi.model.OWLOntology;
import org.semanticweb.owlapi.model.OWLOntologyCreationException;
import org.semanticweb.owlapi.model.OWLOntologyManager;
import uk.ac.manchester.cs.owl.owlapi.mansyntaxrenderer.ManchesterOWLSyntaxOWLObjectRendererImpl;
+import com.clarkparsia.owlapiv3.XSD;
import com.google.common.collect.Sets;
import com.hp.hpl.jena.ontology.OntModel;
+import com.hp.hpl.jena.ontology.OntModelSpec;
import com.hp.hpl.jena.query.ParameterizedSparqlString;
import com.hp.hpl.jena.query.Query;
import com.hp.hpl.jena.query.QueryExecutionFactory;
@@ -64,6 +68,7 @@
import com.hp.hpl.jena.query.QuerySolution;
import com.hp.hpl.jena.query.ResultSet;
import com.hp.hpl.jena.query.Syntax;
+import com.hp.hpl.jena.rdf.model.Literal;
import com.hp.hpl.jena.rdf.model.Model;
import com.hp.hpl.jena.rdf.model.ModelFactory;
import com.hp.hpl.jena.rdf.model.RDFNode;
@@ -190,7 +195,7 @@
List<? extends EvaluatedDescription> learnedClassExpressions = computeMapping(nc, source, target);
if(learnedClassExpressions != null){
mapping.put(nc, learnedClassExpressions);
- //break;
+// break;
}
} catch (Exception e) {
logger.error("Failed for " + nc.getName(), e);
@@ -217,7 +222,7 @@
}
private List<? extends EvaluatedDescription> learnClassExpressions(KnowledgeBase kb, SortedSet<Individual> positiveExamples, boolean posNeg){
- Model fullFragment = null;
+ OntModel fullFragment = null;
try {
//get a sample of the positive examples
@@ -258,8 +263,6 @@
}
}
- AutomaticNegativeExampleFinderSPARQL2 finder = new AutomaticNegativeExampleFinderSPARQL2(kb.getEndpoint());
- //TODO find negative examples
mon.stop();
logger.info("Found " + negativeExamples.size() + " negative examples in " + mon.getLastValue() + "ms.");
}
@@ -276,7 +279,7 @@
logger.info("...done.");
//create fragment consisting of both
- fullFragment = ModelFactory.createDefaultModel();
+ fullFragment = ModelFactory.createOntologyModel(OntModelSpec.RDFS_MEM);
fullFragment.add(positiveFragment);
fullFragment.add(negativeFragment);
@@ -303,6 +306,15 @@
la.setNoisePercentage(25);
la.init();
la.start();
+
+ try {
+ QTL qtl = new QTL(lp, new LocalModelBasedSparqlEndpointKS(fullFragment));
+ qtl.init();
+ qtl.start();
+ System.out.println(qtl.getSPARQLQuery());
+ } catch (LearningProblemUnsupportedException e) {
+ e.printStackTrace();
+ }
logger.info(la.getCurrentlyBestEvaluatedDescription());
@@ -311,7 +323,7 @@
e.printStackTrace();
try {
new File("errors").mkdir();
- fullFragment.write(new FileOutputStream("errors/" + prettyPrint(currentClass) + ".ttl"), "TURTLE", null);
+ fullFragment.write(new FileOutputStream("errors/" + prettyPrint(currentClass) + "_inconsistent.ttl"), "TURTLE", null);
} catch (FileNotFoundException e1) {
e1.printStackTrace();
}
@@ -329,7 +341,7 @@
} catch (OWLOntologyCreationException e) {
e.printStackTrace();
try {
- model.write(new FileOutputStream(prettyPrint(currentClass) + "_conversion_error.ttl"), "TURTLE", null);
+ model.write(new FileOutputStream("errors/" + prettyPrint(currentClass) + "_conversion_error.ttl"), "TURTLE", null);
} catch (FileNotFoundException e1) {
e.printStackTrace();
}
@@ -376,11 +388,11 @@
Statement st = iter.next();
RDFNode object = st.getObject();
if(object.isLiteral()){
- statementsToRemove.add(st);
-// Literal lit = object.asLiteral();
-// if(lit.getDatatype() == null || lit.getDatatype().equals(XSD.STRING)){
-// iter.remove();
-// }
+// statementsToRemove.add(st);
+ Literal lit = object.asLiteral();
+ if(lit.getDatatype() == null || lit.getDatatype().equals(XSD.STRING)){
+ st.changeObject("shortened", "en");
+ }
}
}
fullFragment.remove(statementsToRemove);
Modified: trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java
===================================================================
--- trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java 2013-02-06 18:17:28 UTC (rev 3896)
+++ trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java 2013-02-07 14:03:54 UTC (rev 3897)
@@ -100,7 +100,7 @@
}
@Test
- public void testSingleClass() {
+ public void testSingleClassOpenCycToDBpedia() {
OntologyMatching matcher = new OntologyMatching(openCyc, dbpedia);
NamedClass nc = new NamedClass("http://sw.opencyc.org/concept/Mx4r4fYeXvbPQdiKtoNafhmOew");
List<? extends EvaluatedDescription> mapping = matcher.computeMapping(nc, openCyc, dbpedia);
@@ -109,6 +109,26 @@
System.out.println(OntologyMatching.toHTMLWithLabels(alignment, openCyc, dbpedia));
}
+ @Test
+ public void testSingleClassLinkedGeoDataToDBpedia() {
+ OntologyMatching matcher = new OntologyMatching(linkedGeoData, dbpedia);
+ NamedClass nc = new NamedClass("http://linkedgeodata.org/ontology/Aerodrome");
+ List<? extends EvaluatedDescription> mapping = matcher.computeMapping(nc, linkedGeoData, dbpedia);
+ Map<Description, List<? extends EvaluatedDescription>> alignment = new HashMap<Description, List<? extends EvaluatedDescription>>();
+ alignment.put(nc, mapping);
+ System.out.println(OntologyMatching.toHTMLWithLabels(alignment, linkedGeoData, dbpedia));
+ }
+
+ @Test
+ public void testSingleClassDBpediaToLinkedGeoData() {
+ OntologyMatching matcher = new OntologyMatching(dbpedia, linkedGeoData);
+ NamedClass nc = new NamedClass("http://dbpedia.org/ontology/AdministrativeRegion");
+ List<? extends EvaluatedDescription> mapping = matcher.computeMapping(nc, dbpedia, linkedGeoData);
+ Map<Description, List<? extends EvaluatedDescription>> alignment = new HashMap<Description, List<? extends EvaluatedDescription>>();
+ alignment.put(nc, mapping);
+ System.out.println(OntologyMatching.toHTMLWithLabels(alignment, dbpedia, linkedGeoData));
+ }
+
private void save(String filename, Map<Description, List<? extends EvaluatedDescription>> mapping, KnowledgeBase source, KnowledgeBase target){
BufferedWriter out = null;
try {
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-06 18:17:36
|
Revision: 3896
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3896&view=rev
Author: lorenz_b
Date: 2013-02-06 18:17:28 +0000 (Wed, 06 Feb 2013)
Log Message:
-----------
Moved to Unit test folder.
Added Paths:
-----------
trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java
Removed Paths:
-------------
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java
Deleted: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java 2013-02-06 18:01:42 UTC (rev 3895)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java 2013-02-06 18:17:28 UTC (rev 3896)
@@ -1,147 +0,0 @@
-package org.dllearner.scripts;
-
-import java.io.BufferedWriter;
-import java.io.FileWriter;
-import java.io.IOException;
-import java.net.URL;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-
-import org.apache.log4j.ConsoleAppender;
-import org.apache.log4j.Level;
-import org.apache.log4j.Logger;
-import org.apache.log4j.PatternLayout;
-import org.dllearner.core.EvaluatedDescription;
-import org.dllearner.core.owl.Description;
-import org.dllearner.core.owl.NamedClass;
-import org.dllearner.kb.extraction.ExtractionAlgorithm;
-import org.dllearner.kb.sparql.ExtractionDBCache;
-import org.dllearner.kb.sparql.SparqlEndpoint;
-import org.dllearner.kb.sparql.SparqlKnowledgeSource;
-import org.dllearner.scripts.OntologyMatching.KnowledgeBase;
-import org.junit.Before;
-import org.junit.Test;
-import org.semanticweb.owlapi.io.ToStringRenderer;
-
-import uk.ac.manchester.cs.owl.owlapi.mansyntaxrenderer.ManchesterOWLSyntaxOWLObjectRendererImpl;
-
-public class OntologyMatchingTest {
-
- private KnowledgeBase dbpedia;
- private KnowledgeBase worldFactBook;
- private KnowledgeBase openCyc;
- private KnowledgeBase linkedGeoData;
-
- @Before
- public void setUp() throws Exception {
- // render output
- ToStringRenderer.getInstance().setRenderer(new ManchesterOWLSyntaxOWLObjectRendererImpl());
- // set logging properties
- Logger.getLogger(SparqlKnowledgeSource.class).setLevel(Level.WARN);
- Logger.getLogger(ExtractionAlgorithm.class).setLevel(Level.WARN);
- Logger.getLogger(org.dllearner.kb.extraction.Manager.class).setLevel(Level.WARN);
- Logger.getRootLogger().removeAllAppenders();
- Logger.getRootLogger().addAppender(new ConsoleAppender(new PatternLayout("%m%n")));
-// Logger.getRootLogger().setLevel(Level.DEBUG);
-
- //DBpedia
- SparqlEndpoint endpoint = SparqlEndpoint.getEndpointDBpedia();
- ExtractionDBCache cache = new ExtractionDBCache("cache");
- String namespace = "http://dbpedia.org/resource/";
- dbpedia = new KnowledgeBase(endpoint, null, namespace);
-
- //World Factbook
- //TODO problem with World Factbook is that old FU Berlin server is useless because of bugs and current version
- //is provide by University Of Mannheim now with another namespace http://wifo5-03.informatik.uni-mannheim.de/factbook/resource/
- //but the DBpedia links are still to the old D2R server instance
- //workaround: replace namespace before learning
- endpoint = new SparqlEndpoint(new URL("http://wifo5-03.informatik.uni-mannheim.de/factbook/sparql"));
- cache = new ExtractionDBCache("cache");
- namespace = "http://www4.wiwiss.fu-berlin.de/factbook/resource/";
- worldFactBook = new KnowledgeBase(endpoint, cache, namespace);
-
- //local OpenCyc
- endpoint = new SparqlEndpoint(new URL("http://localhost:8890/sparql"));
- cache = new ExtractionDBCache("cache");
- namespace = "http://sw.cyc.com";
- openCyc = new KnowledgeBase(endpoint, null, namespace);
-
- //LinkedGeoData
- endpoint = new SparqlEndpoint(new URL("http://linkedgeodata.org/sparql"));
- cache = new ExtractionDBCache("cache");
- namespace = "http://linkedgeodata.org/triplify/";
- linkedGeoData = new KnowledgeBase(endpoint, null, namespace);
- }
-
- @Test
- public void testDBpediaWorldFactbook() {
- OntologyMatching matcher = new OntologyMatching(dbpedia, worldFactBook);
- matcher.start();
- save(matcher.getMappingKB1KB2(), "dbpedia_wfb.html");
- save(matcher.getMappingKB2KB1(), "wfb_dbpedia.html");
- }
-
- @Test
- public void testDBpediaOpenCyc() {
- OntologyMatching matcher = new OntologyMatching(dbpedia, openCyc);
- matcher.start();
- save("dbpedia_opencyc.html", matcher.getMappingKB1KB2(), dbpedia, openCyc);
- save("opencyc_dbpedia.html", matcher.getMappingKB2KB1(), openCyc, dbpedia);
- }
-
- @Test
- public void testDBpediaLinkedGeoData() {
- OntologyMatching matcher = new OntologyMatching(dbpedia, linkedGeoData);
- matcher.start();
- save("dbpedia_lgd.html", matcher.getMappingKB1KB2(), dbpedia, linkedGeoData );
- save("lgd_dbpedia.html", matcher.getMappingKB2KB1(), linkedGeoData, dbpedia);
- }
-
- @Test
- public void testSingleClass() {
- OntologyMatching matcher = new OntologyMatching(openCyc, dbpedia);
- NamedClass nc = new NamedClass("http://sw.opencyc.org/concept/Mx4r4fYeXvbPQdiKtoNafhmOew");
- List<? extends EvaluatedDescription> mapping = matcher.computeMapping(nc, openCyc, dbpedia);
- Map<Description, List<? extends EvaluatedDescription>> alignment = new HashMap<Description, List<? extends EvaluatedDescription>>();
- alignment.put(nc, mapping);
- System.out.println(OntologyMatching.toHTMLWithLabels(alignment, openCyc, dbpedia));
- }
-
- private void save(String filename, Map<Description, List<? extends EvaluatedDescription>> mapping, KnowledgeBase source, KnowledgeBase target){
- BufferedWriter out = null;
- try {
- out = new BufferedWriter(new FileWriter(filename));
- out.write(OntologyMatching.toHTMLWithLabels(mapping, source, target));
- } catch (IOException e) {
- e.printStackTrace();
- } finally {
- if (out != null) {
- try {
- out.close();
- } catch (IOException e) {
- e.printStackTrace();
- }
- }
- }
- }
-
- private void save(Map<Description, List<? extends EvaluatedDescription>> mapping, String filename){
- BufferedWriter out = null;
- try {
- out = new BufferedWriter(new FileWriter(filename));
- out.write(OntologyMatching.toHTML(mapping));
- } catch (IOException e) {
- e.printStackTrace();
- } finally {
- if (out != null) {
- try {
- out.close();
- } catch (IOException e) {
- e.printStackTrace();
- }
- }
- }
- }
-
-}
Added: trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java
===================================================================
--- trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java (rev 0)
+++ trunk/scripts/src/test/java/org/dllearner/junit/OntologyMatchingTest.java 2013-02-06 18:17:28 UTC (rev 3896)
@@ -0,0 +1,148 @@
+package org.dllearner.junit;
+
+import java.io.BufferedWriter;
+import java.io.FileWriter;
+import java.io.IOException;
+import java.net.URL;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+
+import org.apache.log4j.ConsoleAppender;
+import org.apache.log4j.Level;
+import org.apache.log4j.Logger;
+import org.apache.log4j.PatternLayout;
+import org.dllearner.core.EvaluatedDescription;
+import org.dllearner.core.owl.Description;
+import org.dllearner.core.owl.NamedClass;
+import org.dllearner.kb.extraction.ExtractionAlgorithm;
+import org.dllearner.kb.sparql.ExtractionDBCache;
+import org.dllearner.kb.sparql.SparqlEndpoint;
+import org.dllearner.kb.sparql.SparqlKnowledgeSource;
+import org.dllearner.scripts.OntologyMatching;
+import org.dllearner.scripts.OntologyMatching.KnowledgeBase;
+import org.junit.Before;
+import org.junit.Test;
+import org.semanticweb.owlapi.io.ToStringRenderer;
+
+import uk.ac.manchester.cs.owl.owlapi.mansyntaxrenderer.ManchesterOWLSyntaxOWLObjectRendererImpl;
+
+public class OntologyMatchingTest {
+
+ private KnowledgeBase dbpedia;
+ private KnowledgeBase worldFactBook;
+ private KnowledgeBase openCyc;
+ private KnowledgeBase linkedGeoData;
+
+ @Before
+ public void setUp() throws Exception {
+ // render output
+ ToStringRenderer.getInstance().setRenderer(new ManchesterOWLSyntaxOWLObjectRendererImpl());
+ // set logging properties
+ Logger.getLogger(SparqlKnowledgeSource.class).setLevel(Level.WARN);
+ Logger.getLogger(ExtractionAlgorithm.class).setLevel(Level.WARN);
+ Logger.getLogger(org.dllearner.kb.extraction.Manager.class).setLevel(Level.WARN);
+ Logger.getRootLogger().removeAllAppenders();
+ Logger.getRootLogger().addAppender(new ConsoleAppender(new PatternLayout("%m%n")));
+// Logger.getRootLogger().setLevel(Level.DEBUG);
+
+ //DBpedia
+ SparqlEndpoint endpoint = SparqlEndpoint.getEndpointDBpedia();
+ ExtractionDBCache cache = new ExtractionDBCache("cache");
+ String namespace = "http://dbpedia.org/resource/";
+ dbpedia = new KnowledgeBase(endpoint, null, namespace);
+
+ //World Factbook
+ //TODO problem with World Factbook is that old FU Berlin server is useless because of bugs and current version
+ //is provide by University Of Mannheim now with another namespace http://wifo5-03.informatik.uni-mannheim.de/factbook/resource/
+ //but the DBpedia links are still to the old D2R server instance
+ //workaround: replace namespace before learning
+ endpoint = new SparqlEndpoint(new URL("http://wifo5-03.informatik.uni-mannheim.de/factbook/sparql"));
+ cache = new ExtractionDBCache("cache");
+ namespace = "http://www4.wiwiss.fu-berlin.de/factbook/resource/";
+ worldFactBook = new KnowledgeBase(endpoint, cache, namespace);
+
+ //local OpenCyc
+ endpoint = new SparqlEndpoint(new URL("http://localhost:8890/sparql"));
+ cache = new ExtractionDBCache("cache");
+ namespace = "http://sw.cyc.com";
+ openCyc = new KnowledgeBase(endpoint, null, namespace);
+
+ //LinkedGeoData
+ endpoint = new SparqlEndpoint(new URL("http://linkedgeodata.org/sparql"));
+ cache = new ExtractionDBCache("cache");
+ namespace = "http://linkedgeodata.org/triplify/";
+ linkedGeoData = new KnowledgeBase(endpoint, null, namespace);
+ }
+
+ @Test
+ public void testDBpediaWorldFactbook() {
+ OntologyMatching matcher = new OntologyMatching(dbpedia, worldFactBook);
+ matcher.start();
+ save(matcher.getMappingKB1KB2(), "dbpedia_wfb.html");
+ save(matcher.getMappingKB2KB1(), "wfb_dbpedia.html");
+ }
+
+ @Test
+ public void testDBpediaOpenCyc() {
+ OntologyMatching matcher = new OntologyMatching(dbpedia, openCyc);
+ matcher.start();
+ save("dbpedia_opencyc.html", matcher.getMappingKB1KB2(), dbpedia, openCyc);
+ save("opencyc_dbpedia.html", matcher.getMappingKB2KB1(), openCyc, dbpedia);
+ }
+
+ @Test
+ public void testDBpediaLinkedGeoData() {
+ OntologyMatching matcher = new OntologyMatching(dbpedia, linkedGeoData);
+ matcher.start();
+ save("dbpedia_lgd.html", matcher.getMappingKB1KB2(), dbpedia, linkedGeoData );
+ save("lgd_dbpedia.html", matcher.getMappingKB2KB1(), linkedGeoData, dbpedia);
+ }
+
+ @Test
+ public void testSingleClass() {
+ OntologyMatching matcher = new OntologyMatching(openCyc, dbpedia);
+ NamedClass nc = new NamedClass("http://sw.opencyc.org/concept/Mx4r4fYeXvbPQdiKtoNafhmOew");
+ List<? extends EvaluatedDescription> mapping = matcher.computeMapping(nc, openCyc, dbpedia);
+ Map<Description, List<? extends EvaluatedDescription>> alignment = new HashMap<Description, List<? extends EvaluatedDescription>>();
+ alignment.put(nc, mapping);
+ System.out.println(OntologyMatching.toHTMLWithLabels(alignment, openCyc, dbpedia));
+ }
+
+ private void save(String filename, Map<Description, List<? extends EvaluatedDescription>> mapping, KnowledgeBase source, KnowledgeBase target){
+ BufferedWriter out = null;
+ try {
+ out = new BufferedWriter(new FileWriter(filename));
+ out.write(OntologyMatching.toHTMLWithLabels(mapping, source, target));
+ } catch (IOException e) {
+ e.printStackTrace();
+ } finally {
+ if (out != null) {
+ try {
+ out.close();
+ } catch (IOException e) {
+ e.printStackTrace();
+ }
+ }
+ }
+ }
+
+ private void save(Map<Description, List<? extends EvaluatedDescription>> mapping, String filename){
+ BufferedWriter out = null;
+ try {
+ out = new BufferedWriter(new FileWriter(filename));
+ out.write(OntologyMatching.toHTML(mapping));
+ } catch (IOException e) {
+ e.printStackTrace();
+ } finally {
+ if (out != null) {
+ try {
+ out.close();
+ } catch (IOException e) {
+ e.printStackTrace();
+ }
+ }
+ }
+ }
+
+}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-06 18:01:57
|
Revision: 3895
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3895&view=rev
Author: lorenz_b
Date: 2013-02-06 18:01:42 +0000 (Wed, 06 Feb 2013)
Log Message:
-----------
Continued matching component.
Modified Paths:
--------------
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-02-06 17:59:21 UTC (rev 3894)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-02-06 18:01:42 UTC (rev 3895)
@@ -1,7 +1,15 @@
package org.dllearner.scripts;
+import java.io.ByteArrayInputStream;
+import java.io.ByteArrayOutputStream;
+import java.io.File;
+import java.io.FileNotFoundException;
+import java.io.FileOutputStream;
import java.text.DecimalFormat;
+import java.util.ArrayList;
import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Iterator;
import java.util.List;
import java.util.Map;
import java.util.Map.Entry;
@@ -15,11 +23,15 @@
import org.dllearner.core.AbstractReasonerComponent;
import org.dllearner.core.ComponentInitException;
import org.dllearner.core.EvaluatedDescription;
+import org.dllearner.core.KnowledgeSource;
import org.dllearner.core.owl.Description;
import org.dllearner.core.owl.Individual;
import org.dllearner.core.owl.NamedClass;
import org.dllearner.core.owl.ObjectProperty;
+import org.dllearner.kb.OWLAPIOntology;
import org.dllearner.kb.SparqlEndpointKS;
+import org.dllearner.kb.sparql.ConciseBoundedDescriptionGenerator;
+import org.dllearner.kb.sparql.ConciseBoundedDescriptionGeneratorImpl;
import org.dllearner.kb.sparql.ExtractionDBCache;
import org.dllearner.kb.sparql.SPARQLTasks;
import org.dllearner.kb.sparql.SparqlEndpoint;
@@ -29,18 +41,33 @@
import org.dllearner.learningproblems.PosOnlyLP;
import org.dllearner.reasoning.FastInstanceChecker;
import org.dllearner.reasoning.SPARQLReasoner;
+import org.dllearner.utilities.LabelShortFormProvider;
import org.dllearner.utilities.datastructures.Datastructures;
import org.dllearner.utilities.datastructures.SetManipulation;
import org.dllearner.utilities.datastructures.SortedSetTuple;
import org.dllearner.utilities.examples.AutomaticNegativeExampleFinderSPARQL2;
import org.dllearner.utilities.owl.OWLAPIDescriptionConvertVisitor;
+import org.semanticweb.owlapi.apibinding.OWLManager;
+import org.semanticweb.owlapi.io.UnparsableOntologyException;
+import org.semanticweb.owlapi.model.OWLOntology;
+import org.semanticweb.owlapi.model.OWLOntologyCreationException;
+import org.semanticweb.owlapi.model.OWLOntologyManager;
+import uk.ac.manchester.cs.owl.owlapi.mansyntaxrenderer.ManchesterOWLSyntaxOWLObjectRendererImpl;
+
+import com.google.common.collect.Sets;
+import com.hp.hpl.jena.ontology.OntModel;
+import com.hp.hpl.jena.query.ParameterizedSparqlString;
import com.hp.hpl.jena.query.Query;
+import com.hp.hpl.jena.query.QueryExecutionFactory;
import com.hp.hpl.jena.query.QueryFactory;
import com.hp.hpl.jena.query.QuerySolution;
import com.hp.hpl.jena.query.ResultSet;
import com.hp.hpl.jena.query.Syntax;
+import com.hp.hpl.jena.rdf.model.Model;
+import com.hp.hpl.jena.rdf.model.ModelFactory;
import com.hp.hpl.jena.rdf.model.RDFNode;
+import com.hp.hpl.jena.rdf.model.Statement;
import com.hp.hpl.jena.sparql.engine.http.QueryEngineHTTP;
import com.hp.hpl.jena.vocabulary.OWL;
import com.jamonapi.Monitor;
@@ -59,6 +86,22 @@
//KB2
private KnowledgeBase kb2;
+ private Map<Description, List<? extends EvaluatedDescription>> mappingKB1KB2;
+ private Map<Description, List<? extends EvaluatedDescription>> mappingKB2KB1;
+
+ private boolean posNegLearning = true;
+
+ /**
+ * The maximum number of positive examples, used for the SPARQL extraction and learning algorithm
+ */
+ private int maxNrOfPositiveExamples = 20;
+ /**
+ * The maximum number of negative examples, used for the SPARQL extraction and learning algorithm
+ */
+ private int maxNrOfNegativeExamples = 20;
+
+ private NamedClass currentClass;
+
public OntologyMatching(KnowledgeBase kb1, KnowledgeBase kb2) {
this.kb1 = kb1;
this.kb2 = kb2;
@@ -71,12 +114,28 @@
}
public void start(){
- Map<Description, List<? extends EvaluatedDescription>> mapping1 = computeMapping(kb1, kb2);
- printMappingPretty(mapping1);
- Map<Description, List<? extends EvaluatedDescription>> mapping2 = computeMapping(kb2, kb1);
- printMappingPretty(mapping2);
+ mappingKB1KB2 = computeAlignment(kb1, kb2);
+ printMappingPretty(mappingKB1KB2);
+ mappingKB2KB1 = computeAlignment(kb2, kb1);
+ printMappingPretty(mappingKB2KB1);
}
+ public Map<Description, List<? extends EvaluatedDescription>> getMappingKB1KB2() {
+ return mappingKB1KB2;
+ }
+
+ public Map<Description, List<? extends EvaluatedDescription>> getMappingKB2KB1() {
+ return mappingKB2KB1;
+ }
+
+ public void setMaxNrOfPositiveExamples(int maxNrOfPositiveExamples) {
+ this.maxNrOfPositiveExamples = maxNrOfPositiveExamples;
+ }
+
+ public void setMaxNrOfNegativeExamples(int maxNrOfNegativeExamples) {
+ this.maxNrOfNegativeExamples = maxNrOfNegativeExamples;
+ }
+
private void printMapping(Map<Description, List<? extends EvaluatedDescription>> mapping){
logger.info("Source Class -> Target Class Expression");
for (Entry<Description, List<? extends org.dllearner.core.EvaluatedDescription>> entry : mapping.entrySet()) {
@@ -106,78 +165,139 @@
}
List<? extends org.dllearner.core.EvaluatedDescription> value = entry.getValue();
logger.info(OWLAPIDescriptionConvertVisitor.getOWLClassExpression(key));
- for (EvaluatedDescription evaluatedDescription : value) {
- logger.info(indention + "\t->\t" +
- OWLAPIDescriptionConvertVisitor.getOWLClassExpression(evaluatedDescription.getDescription()) +
- "(" + dfPercent.format(evaluatedDescription.getAccuracy()) + ")");
+ if(value == null){
+ logger.info(indention + "\t->\t ERROR");
+ } else {
+ for (EvaluatedDescription evaluatedDescription : value) {
+ logger.info(indention + "\t->\t" +
+ OWLAPIDescriptionConvertVisitor.getOWLClassExpression(evaluatedDescription.getDescription()) +
+ "(" + dfPercent.format(evaluatedDescription.getAccuracy()) + ")");
+ }
}
+
}
}
- private Map<Description, List<? extends EvaluatedDescription>> computeMapping(KnowledgeBase source, KnowledgeBase target) {
+ private Map<Description, List<? extends EvaluatedDescription>> computeAlignment(KnowledgeBase source, KnowledgeBase target) {
Map<Description, List<? extends EvaluatedDescription>> mapping = new HashMap<Description, List<? extends EvaluatedDescription>>();
// get all classes in SOURCE
Set<NamedClass> sourceClasses = getClasses(source);
// for each class of KB1
for (NamedClass nc : sourceClasses) {
- logger.info(nc);
- // get all via owl:sameAs related individuals
- SortedSet<Individual> individuals = getRelatedIndividualsNamespaceAware(source, nc, target.getNamespace());
- logger.info(individuals.size());
- //learn concept in KB2 based on the examples
- if(individuals.size() >= 3){
- individuals = SetManipulation.fuzzyShrinkInd(individuals, 10);
- List<? extends EvaluatedDescription> learnedClassExpressions = learnClassExpression(target, individuals);
- mapping.put(nc, learnedClassExpressions);
+ try {
+ logger.info(nc);
+ List<? extends EvaluatedDescription> learnedClassExpressions = computeMapping(nc, source, target);
+ if(learnedClassExpressions != null){
+ mapping.put(nc, learnedClassExpressions);
+ //break;
+ }
+ } catch (Exception e) {
+ logger.error("Failed for " + nc.getName(), e);
}
}
return mapping;
}
- private List<? extends EvaluatedDescription> learnClassExpression(KnowledgeBase kb, SortedSet<Individual> posExamples){
- return learnClassExpression(kb, posExamples, false);
+ public List<? extends EvaluatedDescription> computeMapping(NamedClass sourceClass, KnowledgeBase source, KnowledgeBase target){
+ currentClass = sourceClass;
+ List<? extends EvaluatedDescription> learnedClassExpressions = null;
+ // get all via owl:sameAs related individuals
+ SortedSet<Individual> relatedIndividuals = getRelatedIndividualsNamespaceAware(source, sourceClass, target.getNamespace());
+ logger.info("#Resources in target KB: " + relatedIndividuals.size());
+ //learn concept in KB2 based on the examples
+ if(relatedIndividuals.size() >= 3){
+ learnedClassExpressions = learnClassExpressions(target, relatedIndividuals, posNegLearning);
+ }
+ return learnedClassExpressions;
}
- private List<? extends EvaluatedDescription> learnClassExpression(KnowledgeBase kb, SortedSet<Individual> positiveExamples, boolean posNeg){
+ private List<? extends EvaluatedDescription> learnClassExpressions(KnowledgeBase kb, SortedSet<Individual> posExamples){
+ return learnClassExpressions(kb, posExamples, false);
+ }
+
+ private List<? extends EvaluatedDescription> learnClassExpressions(KnowledgeBase kb, SortedSet<Individual> positiveExamples, boolean posNeg){
+ Model fullFragment = null;
try {
+
+ //get a sample of the positive examples
+ SortedSet<Individual> positiveExamplesSample = SetManipulation.stableShrinkInd(positiveExamples, maxNrOfPositiveExamples);
+
+ //starting from the positive examples, we first extract the fragment for them
+ logger.info("Extracting fragment for positive examples...");
+ Model positiveFragment = getFragment(positiveExamplesSample, kb);
+ logger.info("...done.");
+
+ //based on the fragment we try to find some good negative examples
SortedSet<Individual> negativeExamples = new TreeSet<Individual>();
if(posNeg){
- //find negative examples
mon.start();
+ //find the classes the positive examples are asserted to
+ Set<NamedClass> classes = new HashSet<NamedClass>();
+ ParameterizedSparqlString template = new ParameterizedSparqlString("SELECT ?type WHERE {?s a ?type.}");
+ for(Individual pos : positiveExamples){
+ template.clearParams();
+ template.setIri("s", pos.getName());
+ ResultSet rs = QueryExecutionFactory.create(template.asQuery(), positiveFragment).execSelect();
+ QuerySolution qs;
+ while(rs.hasNext()){
+ qs = rs.next();
+ if(qs.get("type").isURIResource()){
+ classes.add(new NamedClass(qs.getResource("type").getURI()));
+ }
+ }
+ }
+ System.out.println(classes);
+
+ //get the negative examples
+ for(NamedClass nc : classes){
+ Set<String> parallelClasses = kb.getSparqlHelper().getParallelClasses(nc.getName(), 5);
+ for(String parallelClass : parallelClasses){
+ negativeExamples.addAll(kb.getReasoner().getIndividuals(new NamedClass(parallelClass), 5));
+ negativeExamples.removeAll(positiveExamples);
+ }
+ }
+
AutomaticNegativeExampleFinderSPARQL2 finder = new AutomaticNegativeExampleFinderSPARQL2(kb.getEndpoint());
//TODO find negative examples
mon.stop();
logger.info("Found " + negativeExamples.size() + " negative examples in " + mon.getLastValue() + "ms.");
}
- SortedSetTuple<Individual> examples = new SortedSetTuple<Individual>(positiveExamples, negativeExamples);
+ //get a sample of the positive examples
+ SortedSet<Individual> negativeExamplesSample = SetManipulation.stableShrinkInd(negativeExamples, maxNrOfNegativeExamples);
- SparqlKnowledgeSource ks = new SparqlKnowledgeSource();
- ks.setInstances(Datastructures.individualSetToStringSet(examples.getCompleteSet()));
- ks.setUrl(kb.getEndpoint().getURL());
- ks.setDefaultGraphURIs(new TreeSet<String>(kb.getEndpoint().getDefaultGraphURIs()));
- ks.setUseLits(false);
- ks.setUseCacheDatabase(true);
- ks.setCacheDir("cache");
- ks.setRecursionDepth(2);
- ks.setCloseAfterRecursion(true);
- ks.setDissolveBlankNodes(false);
- ks.setSaveExtractedFragment(false);
- ks.init();
+ logger.info("#Positive examples: " + positiveExamplesSample.size());
+ logger.info("#Negative examples: " + negativeExamplesSample.size());
+ //create fragment for negative examples
+ logger.info("Extracting fragment for negative examples...");
+ Model negativeFragment = getFragment(negativeExamplesSample, kb);
+ logger.info("...done.");
+
+ //create fragment consisting of both
+ fullFragment = ModelFactory.createDefaultModel();
+ fullFragment.add(positiveFragment);
+ fullFragment.add(negativeFragment);
+
+ //here is the most difficult task, i.e. find a 'good' fragment of the KB on which we can learn
+ KnowledgeSource ks = convert(fullFragment);
+
+ //initialize the reasoner
AbstractReasonerComponent rc = new FastInstanceChecker(ks);
rc.init();
+ //initialize the learning problem
AbstractLearningProblem lp;
if(posNeg){
- lp = new PosNegLPStandard(rc, positiveExamples, negativeExamples);
+ lp = new PosNegLPStandard(rc, positiveExamplesSample, negativeExamplesSample);
} else {
- lp = new PosOnlyLP(rc, positiveExamples);
-
+ lp = new PosOnlyLP(rc, positiveExamplesSample);
}
lp.init();
+ //apply the learning algorithm
+ logger.info("Running learning algorithm...");
CELOE la = new CELOE(lp, rc);
la.setMaxExecutionTimeInSeconds(10);
la.setNoisePercentage(25);
@@ -189,10 +309,104 @@
return la.getCurrentlyBestEvaluatedDescriptions(10);
} catch (ComponentInitException e) {
e.printStackTrace();
+ try {
+ new File("errors").mkdir();
+ fullFragment.write(new FileOutputStream("errors/" + prettyPrint(currentClass) + ".ttl"), "TURTLE", null);
+ } catch (FileNotFoundException e1) {
+ e1.printStackTrace();
+ }
}
return null;
}
+ private KnowledgeSource convert(Model model){
+ try {
+ ByteArrayOutputStream baos = new ByteArrayOutputStream();
+ model.write(baos, "TURTLE", null);
+ OWLOntologyManager man = OWLManager.createOWLOntologyManager();
+ OWLOntology ontology = man.loadOntologyFromOntologyDocument(new ByteArrayInputStream(baos.toByteArray()));
+ return new OWLAPIOntology(ontology);
+ } catch (OWLOntologyCreationException e) {
+ e.printStackTrace();
+ try {
+ model.write(new FileOutputStream(prettyPrint(currentClass) + "_conversion_error.ttl"), "TURTLE", null);
+ } catch (FileNotFoundException e1) {
+ e.printStackTrace();
+ }
+ }
+ return null;
+ }
+
+ /**
+ * Computes a fragment containing hopefully useful information about the resources.
+ * @param ind
+ */
+ private KnowledgeSource getFragmentLegacy(SortedSet<Individual> positiveExamples, SortedSet<Individual> negativeExamples, SparqlEndpoint endpoint){
+ SortedSetTuple<Individual> examples = new SortedSetTuple<Individual>(positiveExamples, negativeExamples);
+ SparqlKnowledgeSource ks = new SparqlKnowledgeSource();
+ ks.setInstances(Datastructures.individualSetToStringSet(examples.getCompleteSet()));
+ ks.setUrl(endpoint.getURL());
+ ks.setDefaultGraphURIs(new TreeSet<String>(endpoint.getDefaultGraphURIs()));
+ ks.setUseLits(false);
+ ks.setUseCacheDatabase(true);
+ ks.setCacheDir("cache");
+ ks.setRecursionDepth(2);
+ ks.setCloseAfterRecursion(true);
+ ks.setDissolveBlankNodes(false);
+ ks.setSaveExtractedFragment(false);
+ ks.init();
+ return ks;
+ }
+
+ /**
+ * Computes a fragment containing hopefully useful information about the resources.
+ * @param ind
+ */
+ private Model getFragment(SortedSet<Individual> positiveExamples, SortedSet<Individual> negativeExamples, KnowledgeBase kb){
+ OntModel fullFragment = ModelFactory.createOntologyModel();
+ int i = 1;
+ int size = Sets.union(positiveExamples, negativeExamples).size();
+ for(Individual ind : Sets.union(positiveExamples, negativeExamples)){
+ logger.info(i++ + "/" + size);
+ fullFragment.add(getFragment(ind, kb));
+ }
+ //filter out triples with String literals, as there often occur are some syntax errors and they are not relevant for learning
+ List<Statement> statementsToRemove = new ArrayList<Statement>();
+ for(Iterator<Statement> iter = fullFragment.listStatements().toList().iterator(); iter.hasNext();){
+ Statement st = iter.next();
+ RDFNode object = st.getObject();
+ if(object.isLiteral()){
+ statementsToRemove.add(st);
+// Literal lit = object.asLiteral();
+// if(lit.getDatatype() == null || lit.getDatatype().equals(XSD.STRING)){
+// iter.remove();
+// }
+ }
+ }
+ fullFragment.remove(statementsToRemove);
+ return fullFragment;
+ }
+
+ /**
+ * Computes a fragment containing hopefully useful information about the resources.
+ * @param ind
+ */
+ private Model getFragment(SortedSet<Individual> examples, KnowledgeBase kb){
+ return getFragment(examples, new TreeSet<Individual>(), kb);
+ }
+
+ /**
+ * Computes a fragment containing hopefully useful information about the resource.
+ * @param ind
+ */
+ private Model getFragment(Individual ind, KnowledgeBase kb){
+ logger.debug("Loading fragment for " + ind.getName());
+ ConciseBoundedDescriptionGenerator cbdGen = new ConciseBoundedDescriptionGeneratorImpl(kb.getEndpoint(), kb.getCache());
+ Model cbd = cbdGen.getConciseBoundedDescription(ind.getName(), 2);
+ logger.debug("Got " + cbd.size() + " triples.");
+ return cbd;
+ }
+
private Set<NamedClass> getClasses(KnowledgeBase kb){
Set<NamedClass> classes = kb.getSparqlHelper().getAllClasses();
//fallback: check for ?s a ?type where ?type is not asserted to owl:Class
@@ -282,6 +496,84 @@
return rs;
}
+ private String prettyPrint(Description desc){
+ return OWLAPIDescriptionConvertVisitor.getOWLClassExpression(desc).toString();
+ }
+
+ public static String toHTML(Map<Description, List<? extends EvaluatedDescription>> mapping){
+ StringBuilder sb = new StringBuilder();
+ DecimalFormat dfPercent = new DecimalFormat("0.00%");
+ sb.append("<html>\n");
+ sb.append("<table>\n");
+ sb.append("<thead><tr><th>Source Class</th><th>Target Class Expressions</th></tr></thead>\n");
+ sb.append("<tbody>\n");
+
+
+ for (Entry<Description, List<? extends org.dllearner.core.EvaluatedDescription>> entry : mapping.entrySet()) {
+ Description key = entry.getKey();
+ List<? extends org.dllearner.core.EvaluatedDescription> value = entry.getValue();
+ if(value == null){
+ sb.append("<tr><th>" + OWLAPIDescriptionConvertVisitor.getOWLClassExpression(key) + "</th>\n");
+ sb.append("<tr><td>ERROR</td></tr>\n");
+ } else {
+ sb.append("<tr><th rowspan=\"" + value.size()+1 + "\">" + OWLAPIDescriptionConvertVisitor.getOWLClassExpression(key) + "</th>\n");
+ for (EvaluatedDescription evaluatedDescription : value) {
+ sb.append("<tr><td>" +
+ OWLAPIDescriptionConvertVisitor.getOWLClassExpression(evaluatedDescription.getDescription()) + "(" + dfPercent.format(evaluatedDescription.getAccuracy()) + ")"
+ + "</td></tr>\n");
+ }
+ }
+ }
+
+ sb.append("</tbody>\n");
+ sb.append("</table>\n");
+ sb.append("</html>\n");
+
+ return sb.toString();
+ }
+
+ public static String toHTMLWithLabels(Map<Description, List<? extends EvaluatedDescription>> mapping, KnowledgeBase source, KnowledgeBase target){
+ ManchesterOWLSyntaxOWLObjectRendererImpl sourceRenderer = new ManchesterOWLSyntaxOWLObjectRendererImpl();
+ sourceRenderer.setShortFormProvider(new LabelShortFormProvider(source.getEndpoint(), source.getCache()));
+ ManchesterOWLSyntaxOWLObjectRendererImpl targetRenderer = new ManchesterOWLSyntaxOWLObjectRendererImpl();
+ targetRenderer.setShortFormProvider(new LabelShortFormProvider(target.getEndpoint(), target.getCache()));
+
+ StringBuilder sb = new StringBuilder();
+ DecimalFormat dfPercent = new DecimalFormat("0.00%");
+ sb.append("<html>\n");
+ sb.append("<table>\n");
+ sb.append("<thead><tr><th>Source Class</th><th>Target Class Expressions</th><th>Accuracy</th></tr></thead>\n");
+ sb.append("<tbody>\n");
+
+
+ for (Entry<Description, List<? extends org.dllearner.core.EvaluatedDescription>> entry : mapping.entrySet()) {
+ Description key = entry.getKey();
+ String renderedKey = sourceRenderer.render(OWLAPIDescriptionConvertVisitor.getOWLClassExpression(key));
+ List<? extends org.dllearner.core.EvaluatedDescription> value = entry.getValue();
+ if(value == null){
+ sb.append("<tr><th>" + renderedKey + "</th>\n");
+ sb.append("<tr><td>ERROR</td><td></td></tr>\n");
+ } else {
+ sb.append("<tr><th rowspan=\"" + (value.size()+1) + "\">" + renderedKey + "</th>\n");
+ for (EvaluatedDescription evaluatedDescription : value) {
+ sb.append("<tr>");
+ String renderedDesc = targetRenderer.render(OWLAPIDescriptionConvertVisitor.getOWLClassExpression(evaluatedDescription.getDescription()));
+ sb.append("<td>" + renderedDesc + "</td>");
+ sb.append("<td>" + dfPercent.format(evaluatedDescription.getAccuracy()) + "</td>");
+ sb.append("</tr>\n");
+ }
+ }
+ }
+
+ sb.append("</tbody>\n");
+ sb.append("</table>\n");
+ sb.append("</html>\n");
+
+ return sb.toString();
+ }
+
+
+
public static class KnowledgeBase{
private SparqlEndpoint endpoint;
private SPARQLReasoner reasoner;
@@ -294,7 +586,7 @@
this.namespace = namespace;
this.cache = cache;
- this.reasoner = new SPARQLReasoner(new SparqlEndpointKS(endpoint));
+ this.reasoner = new SPARQLReasoner(new SparqlEndpointKS(endpoint), cache);
this.sparqlHelper = new SPARQLTasks(endpoint);
}
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java 2013-02-06 17:59:21 UTC (rev 3894)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java 2013-02-06 18:01:42 UTC (rev 3895)
@@ -1,11 +1,20 @@
package org.dllearner.scripts;
+import java.io.BufferedWriter;
+import java.io.FileWriter;
+import java.io.IOException;
import java.net.URL;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
import org.apache.log4j.ConsoleAppender;
import org.apache.log4j.Level;
import org.apache.log4j.Logger;
import org.apache.log4j.PatternLayout;
+import org.dllearner.core.EvaluatedDescription;
+import org.dllearner.core.owl.Description;
+import org.dllearner.core.owl.NamedClass;
import org.dllearner.kb.extraction.ExtractionAlgorithm;
import org.dllearner.kb.sparql.ExtractionDBCache;
import org.dllearner.kb.sparql.SparqlEndpoint;
@@ -34,12 +43,13 @@
Logger.getLogger(org.dllearner.kb.extraction.Manager.class).setLevel(Level.WARN);
Logger.getRootLogger().removeAllAppenders();
Logger.getRootLogger().addAppender(new ConsoleAppender(new PatternLayout("%m%n")));
+// Logger.getRootLogger().setLevel(Level.DEBUG);
//DBpedia
SparqlEndpoint endpoint = SparqlEndpoint.getEndpointDBpedia();
ExtractionDBCache cache = new ExtractionDBCache("cache");
String namespace = "http://dbpedia.org/resource/";
- dbpedia = new KnowledgeBase(endpoint, cache, namespace);
+ dbpedia = new KnowledgeBase(endpoint, null, namespace);
//World Factbook
//TODO problem with World Factbook is that old FU Berlin server is useless because of bugs and current version
@@ -55,31 +65,83 @@
endpoint = new SparqlEndpoint(new URL("http://localhost:8890/sparql"));
cache = new ExtractionDBCache("cache");
namespace = "http://sw.cyc.com";
- openCyc = new KnowledgeBase(endpoint, cache, namespace);
+ openCyc = new KnowledgeBase(endpoint, null, namespace);
//LinkedGeoData
endpoint = new SparqlEndpoint(new URL("http://linkedgeodata.org/sparql"));
cache = new ExtractionDBCache("cache");
namespace = "http://linkedgeodata.org/triplify/";
- linkedGeoData = new KnowledgeBase(endpoint, cache, namespace);
+ linkedGeoData = new KnowledgeBase(endpoint, null, namespace);
}
@Test
public void testDBpediaWorldFactbook() {
OntologyMatching matcher = new OntologyMatching(dbpedia, worldFactBook);
matcher.start();
+ save(matcher.getMappingKB1KB2(), "dbpedia_wfb.html");
+ save(matcher.getMappingKB2KB1(), "wfb_dbpedia.html");
}
@Test
public void testDBpediaOpenCyc() {
OntologyMatching matcher = new OntologyMatching(dbpedia, openCyc);
matcher.start();
+ save("dbpedia_opencyc.html", matcher.getMappingKB1KB2(), dbpedia, openCyc);
+ save("opencyc_dbpedia.html", matcher.getMappingKB2KB1(), openCyc, dbpedia);
}
@Test
public void testDBpediaLinkedGeoData() {
OntologyMatching matcher = new OntologyMatching(dbpedia, linkedGeoData);
matcher.start();
+ save("dbpedia_lgd.html", matcher.getMappingKB1KB2(), dbpedia, linkedGeoData );
+ save("lgd_dbpedia.html", matcher.getMappingKB2KB1(), linkedGeoData, dbpedia);
}
+
+ @Test
+ public void testSingleClass() {
+ OntologyMatching matcher = new OntologyMatching(openCyc, dbpedia);
+ NamedClass nc = new NamedClass("http://sw.opencyc.org/concept/Mx4r4fYeXvbPQdiKtoNafhmOew");
+ List<? extends EvaluatedDescription> mapping = matcher.computeMapping(nc, openCyc, dbpedia);
+ Map<Description, List<? extends EvaluatedDescription>> alignment = new HashMap<Description, List<? extends EvaluatedDescription>>();
+ alignment.put(nc, mapping);
+ System.out.println(OntologyMatching.toHTMLWithLabels(alignment, openCyc, dbpedia));
+ }
+
+ private void save(String filename, Map<Description, List<? extends EvaluatedDescription>> mapping, KnowledgeBase source, KnowledgeBase target){
+ BufferedWriter out = null;
+ try {
+ out = new BufferedWriter(new FileWriter(filename));
+ out.write(OntologyMatching.toHTMLWithLabels(mapping, source, target));
+ } catch (IOException e) {
+ e.printStackTrace();
+ } finally {
+ if (out != null) {
+ try {
+ out.close();
+ } catch (IOException e) {
+ e.printStackTrace();
+ }
+ }
+ }
+ }
+
+ private void save(Map<Description, List<? extends EvaluatedDescription>> mapping, String filename){
+ BufferedWriter out = null;
+ try {
+ out = new BufferedWriter(new FileWriter(filename));
+ out.write(OntologyMatching.toHTML(mapping));
+ } catch (IOException e) {
+ e.printStackTrace();
+ } finally {
+ if (out != null) {
+ try {
+ out.close();
+ } catch (IOException e) {
+ e.printStackTrace();
+ }
+ }
+ }
+ }
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-02-06 17:59:29
|
Revision: 3894
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3894&view=rev
Author: lorenz_b
Date: 2013-02-06 17:59:21 +0000 (Wed, 06 Feb 2013)
Log Message:
-----------
Added short form provider for IRIs based on labels in KB.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/kb/sparql/ConciseBoundedDescriptionGeneratorImpl.java
trunk/components-core/src/main/java/org/dllearner/reasoning/SPARQLReasoner.java
Added Paths:
-----------
trunk/components-core/src/main/java/org/dllearner/kb/sparql/QueryEngineHTTP.java
trunk/components-core/src/main/java/org/dllearner/utilities/LabelShortFormProvider.java
Modified: trunk/components-core/src/main/java/org/dllearner/kb/sparql/ConciseBoundedDescriptionGeneratorImpl.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/kb/sparql/ConciseBoundedDescriptionGeneratorImpl.java 2013-01-30 09:55:24 UTC (rev 3893)
+++ trunk/components-core/src/main/java/org/dllearner/kb/sparql/ConciseBoundedDescriptionGeneratorImpl.java 2013-02-06 17:59:21 UTC (rev 3894)
@@ -11,7 +11,7 @@
import com.hp.hpl.jena.query.QueryExecutionFactory;
import com.hp.hpl.jena.rdf.model.Model;
import com.hp.hpl.jena.rdf.model.ModelFactory;
-import com.hp.hpl.jena.sparql.engine.http.QueryEngineHTTP;
+//import com.hp.hpl.jena.sparql.engine.http.QueryEngineHTTP;
public class ConciseBoundedDescriptionGeneratorImpl implements ConciseBoundedDescriptionGenerator{
@@ -166,7 +166,7 @@
}
for (String ngu : endpoint.getNamedGraphURIs()) {
queryExecution.addNamedGraph(ngu);
- }
+ }
model = queryExecution.execConstruct();
} else {
model = cache.executeConstructQuery(endpoint, query);
Added: trunk/components-core/src/main/java/org/dllearner/kb/sparql/QueryEngineHTTP.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/kb/sparql/QueryEngineHTTP.java (rev 0)
+++ trunk/components-core/src/main/java/org/dllearner/kb/sparql/QueryEngineHTTP.java 2013-02-06 17:59:21 UTC (rev 3894)
@@ -0,0 +1,535 @@
+package org.dllearner.kb.sparql;
+
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+
+import java.io.ByteArrayInputStream;
+import java.io.InputStream;
+import java.util.ArrayList;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import java.util.concurrent.TimeUnit;
+
+import org.openjena.atlas.io.IO;
+import org.openjena.riot.Lang;
+import org.openjena.riot.RiotReader;
+import org.openjena.riot.WebContent;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import com.hp.hpl.jena.graph.Triple;
+import com.hp.hpl.jena.query.ARQ;
+import com.hp.hpl.jena.query.Dataset;
+import com.hp.hpl.jena.query.Query;
+import com.hp.hpl.jena.query.QueryException;
+import com.hp.hpl.jena.query.QueryExecException;
+import com.hp.hpl.jena.query.QueryExecution;
+import com.hp.hpl.jena.query.QuerySolution;
+import com.hp.hpl.jena.query.ResultSet;
+import com.hp.hpl.jena.query.ResultSetFactory;
+import com.hp.hpl.jena.rdf.model.Model;
+import com.hp.hpl.jena.sparql.ARQException;
+import com.hp.hpl.jena.sparql.engine.http.HttpParams;
+import com.hp.hpl.jena.sparql.engine.http.HttpQuery;
+import com.hp.hpl.jena.sparql.engine.http.Params;
+import com.hp.hpl.jena.sparql.graph.GraphFactory;
+import com.hp.hpl.jena.sparql.resultset.CSVInput;
+import com.hp.hpl.jena.sparql.resultset.JSONInput;
+import com.hp.hpl.jena.sparql.resultset.XMLInput;
+import com.hp.hpl.jena.sparql.util.Context;
+import com.hp.hpl.jena.util.FileManager;
+
+
+public class QueryEngineHTTP implements QueryExecution
+{
+ private static Logger log = LoggerFactory.getLogger(QueryEngineHTTP.class) ;
+
+ public static final String QUERY_MIME_TYPE = WebContent.contentTypeSPARQLQuery ; // "application/sparql-query" ;
+ private final Query query ;
+ private final String queryString ;
+ private final String service ;
+ private final Context context ;
+
+ //Params
+ Params params = null ;
+
+ // Protocol
+ List<String> defaultGraphURIs = new ArrayList<String>() ;
+ List<String> namedGraphURIs = new ArrayList<String>() ;
+ private String user = null ;
+ private char[] password = null ;
+
+ private boolean finished = false ;
+
+ //Timeouts
+ private long connectTimeout = 0;
+ private TimeUnit connectTimeoutUnit = TimeUnit.MILLISECONDS;
+ private long readTimeout = 0;
+ private TimeUnit readTimeoutUnit = TimeUnit.MILLISECONDS;
+
+ //Compression Support
+ private boolean allowGZip = true ;
+ private boolean allowDeflate = true;
+
+ //Content Types
+ private String selectContentType = WebContent.contentTypeResultsXML;
+ private String askContentType = WebContent.contentTypeResultsXML;
+ private String modelContentType = WebContent.contentTypeRDFXML;
+ public static String[] supportedSelectContentTypes = new String []
+ {
+ WebContent.contentTypeResultsXML,
+ WebContent.contentTypeResultsJSON,
+ WebContent.contentTypeTextTSV,
+ WebContent.contentTypeTextCSV
+ };
+ public static String[] supportedAskContentTypes = new String []
+ {
+ WebContent.contentTypeResultsXML,
+ WebContent.contentTypeJSON
+ };
+
+ // Releasing HTTP input streams is important. We remember this for SELECT,
+ // and will close when the engine is closed
+ private InputStream retainedConnection = null;
+
+ public QueryEngineHTTP(String serviceURI, Query query)
+ {
+ this(serviceURI, query, query.toString()) ;
+ }
+
+ public QueryEngineHTTP(String serviceURI, String queryString)
+ {
+ this(serviceURI, null, queryString) ;
+ }
+
+ private QueryEngineHTTP(String serviceURI, Query query, String queryString)
+ {
+ this.query = query ;
+ this.queryString = queryString ;
+ this.service = serviceURI ;
+ // Copy the global context to freeze it.
+ this.context = new Context(ARQ.getContext()) ;
+ }
+
+// public void setParams(Params params)
+// { this.params = params ; }
+
+ // Meaning-less
+ @Override
+ public void setFileManager(FileManager fm)
+ { throw new UnsupportedOperationException("FileManagers do not apply to remote query execution") ; }
+
+ @Override
+ public void setInitialBinding(QuerySolution binding)
+ { throw new UnsupportedOperationException("Initial bindings not supported for remote queries, consider using a ParameterizedSparqlString to prepare a query for remote execution") ; }
+
+ public void setInitialBindings(ResultSet table)
+ { throw new UnsupportedOperationException("Initial bindings not supported for remote queries, consider using a ParameterizedSparqlString to prepare a query for remote execution") ; }
+
+ /** @param defaultGraphURIs The defaultGraphURIs to set. */
+ public void setDefaultGraphURIs(List<String> defaultGraphURIs)
+ {
+ this.defaultGraphURIs = defaultGraphURIs ;
+ }
+
+ /** @param namedGraphURIs The namedGraphURIs to set. */
+ public void setNamedGraphURIs(List<String> namedGraphURIs)
+ {
+ this.namedGraphURIs = namedGraphURIs ;
+ }
+
+ /**
+ * Sets whether the HTTP request will specify Accept-Encoding: gzip
+ */
+ public void setAllowGZip(boolean allowed)
+ {
+ allowGZip = allowed;
+ }
+
+ /**
+ * Sets whether the HTTP requests will specify Accept-Encoding: deflate
+ */
+ public void setAllowDeflate(boolean allowed)
+ {
+ allowDeflate = allowed;
+ }
+
+ public void addParam(String field, String value)
+ {
+ if ( params == null )
+ params = new Params() ;
+ params.addParam(field, value) ;
+ }
+
+ /** @param defaultGraph The defaultGraph to add. */
+ public void addDefaultGraph(String defaultGraph)
+ {
+ if ( defaultGraphURIs == null )
+ defaultGraphURIs = new ArrayList<String>() ;
+ defaultGraphURIs.add(defaultGraph) ;
+ }
+
+ /** @param name The URI to add. */
+ public void addNamedGraph(String name)
+ {
+ if ( namedGraphURIs == null )
+ namedGraphURIs = new ArrayList<String>() ;
+ namedGraphURIs.add(name) ;
+ }
+
+ /** Set user and password for basic authentication.
+ * After the request is made (one of the exec calls), the application
+ * can overwrite the password array to remove details of the secret.
+ * @param user
+ * @param password
+ */
+ public void setBasicAuthentication(String user, char[] password)
+ {
+ this.user = user ;
+ this.password = password ;
+ }
+
+ @Override
+ public ResultSet execSelect()
+ {
+ HttpQuery httpQuery = makeHttpQuery() ;
+ httpQuery.setAccept(selectContentType) ;
+ InputStream in = httpQuery.exec() ;
+
+ if ( false )
+ {
+ byte b[] = IO.readWholeFile(in) ;
+ String str = new String(b) ;
+ System.out.println(str) ;
+ in = new ByteArrayInputStream(b) ;
+ }
+
+ retainedConnection = in; // This will be closed on close()
+
+ //TODO: Find a way to auto-detect how to create the ResultSet based on the content type in use
+
+ //Don't assume the endpoint actually gives back the content type we asked for
+ String actualContentType = httpQuery.getContentType();
+
+ //If the server fails to return a Content-Type then we will assume
+ //the server returned the type we asked for
+ if (actualContentType == null || actualContentType.equals(""))
+ {
+ actualContentType = selectContentType;
+ }
+
+ if (actualContentType.equals(WebContent.contentTypeResultsXML))
+ return ResultSetFactory.fromXML(in);
+ if (actualContentType.equals(WebContent.contentTypeResultsJSON))
+ return ResultSetFactory.fromJSON(in);
+ if (actualContentType.equals(WebContent.contentTypeTextTSV))
+ return ResultSetFactory.fromTSV(in);
+ if (actualContentType.equals(WebContent.contentTypeTextCSV))
+ return CSVInput.fromCSV(in);
+ throw new QueryException("Endpoint returned Content-Type: " + actualContentType + " which is not currently supported for SELECT queries");
+ }
+
+ @Override
+ public Model execConstruct() { return execConstruct(GraphFactory.makeJenaDefaultModel()) ; }
+
+ @Override
+ public Model execConstruct(Model model) { return execModel(model) ; }
+
+ @Override
+ public Iterator<Triple> execConstructTriples() { return execTriples() ; }
+
+ @Override
+ public Model execDescribe() { return execDescribe(GraphFactory.makeJenaDefaultModel()) ; }
+
+ @Override
+ public Model execDescribe(Model model) { return execModel(model) ; }
+
+ @Override
+ public Iterator<Triple> execDescribeTriples() { return execTriples() ; }
+
+ private Model execModel(Model model)
+ {
+ HttpQuery httpQuery = makeHttpQuery() ;
+ httpQuery.setAccept(WebContent.contentTypeNTriples) ;
+ InputStream in = httpQuery.exec() ;
+
+ //Don't assume the endpoint actually gives back the content type we asked for
+ String actualContentType = httpQuery.getContentType();
+
+ //If the server fails to return a Content-Type then we will assume
+ //the server returned the type we asked for
+ if (actualContentType == null || actualContentType.equals(""))
+ {
+ actualContentType = modelContentType;
+ }
+
+ //Try to select language appropriately here based on the model content type
+ Lang lang = WebContent.contentTypeToLang(actualContentType);
+ if (!lang.isTriples()) throw new QueryException("Endpoint returned Content Type: " + actualContentType + " which is not a valid RDF Graph syntax");
+ model.read(in, null, "N-TRIPLES") ;
+
+ return model ;
+ }
+
+ private Iterator<Triple> execTriples()
+ {
+ HttpQuery httpQuery = makeHttpQuery() ;
+ httpQuery.setAccept(modelContentType) ;
+ InputStream in = httpQuery.exec() ;
+
+ //Don't assume the endpoint actually gives back the content type we asked for
+ String actualContentType = httpQuery.getContentType();
+
+ //If the server fails to return a Content-Type then we will assume
+ //the server returned the type we asked for
+ if (actualContentType == null || actualContentType.equals(""))
+ {
+ actualContentType = modelContentType;
+ }
+
+ //Try to select language appropriately here based on the model content type
+ Lang lang = WebContent.contentTypeToLang(actualContentType);
+ if (!lang.isTriples()) throw new QueryException("Endpoint returned Content Type: " + actualContentType + " which is not a valid RDF Graph syntax");
+
+ return RiotReader.createIteratorTriples(in, lang, null);
+ }
+
+ @Override
+ public boolean execAsk()
+ {
+ HttpQuery httpQuery = makeHttpQuery() ;
+ httpQuery.setAccept(askContentType) ;
+ InputStream in = httpQuery.exec() ;
+
+ try {
+ //Don't assume the endpoint actually gives back the content type we asked for
+ String actualContentType = httpQuery.getContentType();
+
+ //If the server fails to return a Content-Type then we will assume
+ //the server returned the type we asked for
+ if (actualContentType == null || actualContentType.equals(""))
+ {
+ actualContentType = askContentType;
+ }
+
+ //Parse the result appropriately depending on the selected content type
+ if (askContentType.equals(WebContent.contentTypeResultsXML))
+ return XMLInput.booleanFromXML(in) ;
+ if (askContentType.equals(WebContent.contentTypeResultsJSON))
+ return JSONInput.booleanFromJSON(in) ;
+ throw new QueryException("Endpoint returned Content-Type: " + actualContentType + " which is not currently supported for ASK queries");
+ } finally {
+ // Ensure connection is released
+ try { in.close(); }
+ catch (java.io.IOException e) { log.warn("Failed to close connection", e); }
+ }
+ }
+
+ @Override
+ public Context getContext() { return context ; }
+
+ @Override public Dataset getDataset() { return null ; }
+
+ // This may be null - if we were created form a query string,
+ // we don't guarantee to parse it so we let through non-SPARQL
+ // extensions to the far end.
+ @Override public Query getQuery() { return query ; }
+
+ @Override
+ public void setTimeout(long readTimeout)
+ {
+ this.readTimeout = readTimeout;
+ this.readTimeoutUnit = TimeUnit.MILLISECONDS;
+ }
+
+ @Override
+ public void setTimeout(long readTimeout, long connectTimeout)
+ {
+ this.readTimeout = readTimeout;
+ this.readTimeoutUnit = TimeUnit.MILLISECONDS;
+ this.connectTimeout = connectTimeout;
+ this.connectTimeoutUnit = TimeUnit.MILLISECONDS;
+ }
+
+
+ @Override
+ public void setTimeout(long readTimeout, TimeUnit timeoutUnits)
+ {
+ this.readTimeout = readTimeout;
+ this.readTimeoutUnit = timeoutUnits;
+ }
+
+ @Override
+ public void setTimeout(long timeout1, TimeUnit timeUnit1, long timeout2, TimeUnit timeUnit2)
+ {
+ this.readTimeout = timeout1;
+ this.readTimeoutUnit = timeUnit1;
+ this.connectTimeout = timeout2;
+ this.connectTimeoutUnit = timeUnit2;
+ }
+
+ private HttpQuery makeHttpQuery()
+ {
+ // Also need to tie to ResultSet returned which is streamed back if StAX.
+ if ( finished )
+ throw new ARQException("HTTP execution already closed") ;
+
+ HttpQuery httpQuery = new HttpQuery(service) ;
+ httpQuery.merge(getServiceParams(service, context)) ;
+ httpQuery.addParam(HttpParams.pQuery, queryString );
+
+ for ( Iterator<String> iter = defaultGraphURIs.iterator() ; iter.hasNext() ; )
+ {
+ String dft = iter.next() ;
+ httpQuery.addParam(HttpParams.pDefaultGraph, dft) ;
+ }
+ for ( Iterator<String> iter = namedGraphURIs.iterator() ; iter.hasNext() ; )
+ {
+ String name = iter.next() ;
+ httpQuery.addParam(HttpParams.pNamedGraph, name) ;
+ }
+
+ if ( params != null )
+ httpQuery.merge(params) ;
+
+ if (allowGZip)
+ httpQuery.setAllowGZip(true);
+
+ if (allowDeflate)
+ httpQuery.setAllowDeflate(true);
+
+ httpQuery.setBasicAuthentication(user, password) ;
+
+ //Apply timeouts
+ if (connectTimeout > 0)
+ {
+ httpQuery.setConnectTimeout((int)connectTimeoutUnit.toMillis(connectTimeout));
+ }
+ if (readTimeout > 0)
+ {
+ httpQuery.setReadTimeout((int)readTimeoutUnit.toMillis(readTimeout));
+ }
+
+ return httpQuery ;
+ }
+
+
+ // This is to allow setting additional/optional query parameters on a per SERVICE level, see: JENA-195
+ protected static Params getServiceParams(String serviceURI, Context context) throws QueryExecException
+ {
+ Params params = new Params();
+ @SuppressWarnings("unchecked")
+ Map<String, Map<String,List<String>>> serviceParams = (Map<String, Map<String,List<String>>>)context.get(ARQ.serviceParams) ;
+ if ( serviceParams != null )
+ {
+ Map<String,List<String>> paramsMap = serviceParams.get(serviceURI) ;
+ if ( paramsMap != null )
+ {
+ for (String param : paramsMap.keySet())
+ {
+ if ( HttpParams.pQuery.equals(param) )
+ throw new QueryExecException("ARQ serviceParams overrides the 'query' SPARQL protocol parameter") ;
+
+ List<String> values = paramsMap.get(param) ;
+ for (String value : values)
+ params.addParam(param, value) ;
+ }
+ }
+ }
+ return params;
+ }
+
+ public void cancel() { finished = true ; }
+
+ @Override
+ public void abort() { try { close() ; } catch (Exception ex) {} }
+
+ @Override
+ public void close() {
+ finished = false ;
+ if (retainedConnection != null) {
+ try { retainedConnection.close(); }
+ catch (java.io.IOException e) { log.warn("Failed to close connection", e); }
+ finally { retainedConnection = null; }
+ }
+ }
+
+// public boolean isActive() { return false ; }
+
+ @Override
+ public String toString()
+ {
+ HttpQuery httpQuery = makeHttpQuery() ;
+ return "GET "+httpQuery.toString() ;
+ }
+
+ /**
+ * Sets the Content Type for SELECT queries provided that the format is supported
+ * @param contentType
+ */
+ public void setSelectContentType(String contentType)
+ {
+ boolean ok = false;
+ for (String supportedType : supportedSelectContentTypes)
+ {
+ if (supportedType.equals(contentType))
+ {
+ ok = true;
+ break;
+ }
+ }
+ if (!ok) throw new IllegalArgumentException("Given Content Type '" + contentType + "' is not a supported SELECT results format");
+ selectContentType = contentType;
+ }
+
+ /**
+ * Sets the Content Type for ASK queries provided that the format is supported
+ * @param contentType
+ */
+ public void setAskContentType(String contentType)
+ {
+ boolean ok = false;
+ for (String supportedType : supportedAskContentTypes)
+ {
+ if (supportedType.equals(contentType))
+ {
+ ok = true;
+ break;
+ }
+ }
+ if (!ok) throw new IllegalArgumentException("Given Content Type '" + contentType + "' is not a supported ASK results format");
+ askContentType = contentType;
+ }
+
+ /**
+ * Sets the Content Type for CONSTRUCT/DESCRIBE queries provided that the format is supported
+ * @param contentType
+ */
+ public void setModelContentType(String contentType)
+ {
+ //Check that this is a valid setting
+ Lang lang = WebContent.contentTypeToLang(contentType);
+ if (lang == null) throw new IllegalArgumentException("Given Content Type '" + contentType + "' is not supported by RIOT");
+ if (!lang.isTriples()) throw new IllegalArgumentException("Given Content Type '" + contentType + " is not a RDF Graph format");
+
+ modelContentType = contentType;
+ }
+}
+
Modified: trunk/components-core/src/main/java/org/dllearner/reasoning/SPARQLReasoner.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/reasoning/SPARQLReasoner.java 2013-01-30 09:55:24 UTC (rev 3893)
+++ trunk/components-core/src/main/java/org/dllearner/reasoning/SPARQLReasoner.java 2013-02-06 17:59:21 UTC (rev 3894)
@@ -1107,7 +1107,7 @@
logger.debug("Sending query \n {}", query);
ResultSet rs = null;
if(ks.isRemote()){
- if(useCache){
+ if(useCache && cache != null){
rs = SparqlQuery.convertJSONtoResultSet(cache.executeSelectQuery(ks.getEndpoint(), query));
} else {
QueryEngineHTTP queryExecution = new QueryEngineHTTP(ks.getEndpoint().getURL().toString(), query);
Added: trunk/components-core/src/main/java/org/dllearner/utilities/LabelShortFormProvider.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/utilities/LabelShortFormProvider.java (rev 0)
+++ trunk/components-core/src/main/java/org/dllearner/utilities/LabelShortFormProvider.java 2013-02-06 17:59:21 UTC (rev 3894)
@@ -0,0 +1,76 @@
+package org.dllearner.utilities;
+
+import org.dllearner.kb.sparql.ExtractionDBCache;
+import org.dllearner.kb.sparql.SparqlEndpoint;
+import org.dllearner.kb.sparql.SparqlQuery;
+import org.semanticweb.owlapi.model.OWLEntity;
+import org.semanticweb.owlapi.util.ShortFormProvider;
+import org.semanticweb.owlapi.util.SimpleIRIShortFormProvider;
+
+import com.hp.hpl.jena.query.ParameterizedSparqlString;
+import com.hp.hpl.jena.query.Query;
+import com.hp.hpl.jena.query.ResultSet;
+import com.hp.hpl.jena.sparql.engine.http.QueryEngineHTTP;
+import com.hp.hpl.jena.vocabulary.RDFS;
+
+public class LabelShortFormProvider implements ShortFormProvider{
+
+ private final ParameterizedSparqlString queryTemplate = new ParameterizedSparqlString(
+ "SELECT ?label WHERE {?entity ?labelProperty ?label. FILTER(LANGMATCHES(LANG(?label),'en'))} LIMIT 1");
+ private String labelProperty = RDFS.label.getURI();
+
+ private final SimpleIRIShortFormProvider fallback = new SimpleIRIShortFormProvider();
+
+ private ExtractionDBCache cache;
+ private SparqlEndpoint endpoint;
+
+ public LabelShortFormProvider(SparqlEndpoint endpoint) {
+ this.endpoint = endpoint;
+ }
+
+ public LabelShortFormProvider(SparqlEndpoint endpoint, ExtractionDBCache cache) {
+ this.endpoint = endpoint;
+ this.cache = cache;
+ }
+
+ public void setLabelProperty(String labelProperty) {
+ this.labelProperty = labelProperty;
+ }
+
+ @Override
+ public void dispose() {
+ }
+
+ @Override
+ public String getShortForm(OWLEntity entity) {
+ queryTemplate.clearParams();
+ queryTemplate.setIri("entity", entity.toStringID());
+ queryTemplate.setIri("labelProperty", labelProperty);
+ Query query = queryTemplate.asQuery();
+ ResultSet rs = executeSelect(query);
+ String label = null;
+ if(rs.hasNext()){
+ label = rs.next().getLiteral("label").asLiteral().getLexicalForm();
+ } else {
+ label = fallback.getShortForm(entity.getIRI());
+ }
+ return label;
+ }
+
+ protected ResultSet executeSelect(Query query){
+ ResultSet rs = null;
+ if(cache != null){
+ rs = SparqlQuery.convertJSONtoResultSet(cache.executeSelectQuery(endpoint, query.toString()));
+ } else {
+ QueryEngineHTTP qe = new QueryEngineHTTP(endpoint.getURL().toString(), query);
+ for(String uri : endpoint.getDefaultGraphURIs()){
+ qe.addDefaultGraph(uri);
+ }
+ rs = qe.execSelect();
+ }
+ return rs;
+ }
+
+
+
+}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-01-30 09:55:31
|
Revision: 3893
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3893&view=rev
Author: lorenz_b
Date: 2013-01-30 09:55:24 +0000 (Wed, 30 Jan 2013)
Log Message:
-----------
Continued matching.
Modified Paths:
--------------
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
Added Paths:
-----------
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-01-25 13:04:28 UTC (rev 3892)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-01-30 09:55:24 UTC (rev 3893)
@@ -1,6 +1,5 @@
package org.dllearner.scripts;
-import java.net.URL;
import java.text.DecimalFormat;
import java.util.HashMap;
import java.util.List;
@@ -10,10 +9,7 @@
import java.util.SortedSet;
import java.util.TreeSet;
-import org.apache.log4j.ConsoleAppender;
-import org.apache.log4j.Level;
import org.apache.log4j.Logger;
-import org.apache.log4j.PatternLayout;
import org.dllearner.algorithms.celoe.CELOE;
import org.dllearner.core.AbstractLearningProblem;
import org.dllearner.core.AbstractReasonerComponent;
@@ -24,7 +20,6 @@
import org.dllearner.core.owl.NamedClass;
import org.dllearner.core.owl.ObjectProperty;
import org.dllearner.kb.SparqlEndpointKS;
-import org.dllearner.kb.extraction.ExtractionAlgorithm;
import org.dllearner.kb.sparql.ExtractionDBCache;
import org.dllearner.kb.sparql.SPARQLTasks;
import org.dllearner.kb.sparql.SparqlEndpoint;
@@ -35,14 +30,11 @@
import org.dllearner.reasoning.FastInstanceChecker;
import org.dllearner.reasoning.SPARQLReasoner;
import org.dllearner.utilities.datastructures.Datastructures;
+import org.dllearner.utilities.datastructures.SetManipulation;
import org.dllearner.utilities.datastructures.SortedSetTuple;
import org.dllearner.utilities.examples.AutomaticNegativeExampleFinderSPARQL2;
-import org.dllearner.utilities.owl.DLLearnerDescriptionConvertVisitor;
import org.dllearner.utilities.owl.OWLAPIDescriptionConvertVisitor;
-import org.semanticweb.owlapi.io.ToStringRenderer;
-import uk.ac.manchester.cs.owl.owlapi.mansyntaxrenderer.ManchesterOWLSyntaxOWLObjectRendererImpl;
-
import com.hp.hpl.jena.query.Query;
import com.hp.hpl.jena.query.QueryFactory;
import com.hp.hpl.jena.query.QuerySolution;
@@ -132,9 +124,10 @@
logger.info(nc);
// get all via owl:sameAs related individuals
SortedSet<Individual> individuals = getRelatedIndividualsNamespaceAware(source, nc, target.getNamespace());
- logger.info(individuals);
+ logger.info(individuals.size());
//learn concept in KB2 based on the examples
if(individuals.size() >= 3){
+ individuals = SetManipulation.fuzzyShrinkInd(individuals, 10);
List<? extends EvaluatedDescription> learnedClassExpressions = learnClassExpression(target, individuals);
mapping.put(nc, learnedClassExpressions);
}
@@ -260,6 +253,9 @@
String uri = object.asResource().getURI();
//workaround for World Factbook - should be removed later
uri = uri.replace("http://www4.wiwiss.fu-berlin.de/factbook/resource/", "http://wifo5-03.informatik.uni-mannheim.de/factbook/resource/");
+ //workaround for OpenCyc - should be removed later
+ uri = uri.replace("http://sw.cyc.com", "http://sw.opencyc.org");
+
relatedIndividuals.add(new Individual(uri));
}
}
@@ -329,40 +325,5 @@
return cache;
}
-
-
}
-
- /**
- * @param args
- */
- public static void main(String[] args) throws Exception{
- //render output
- ToStringRenderer.getInstance().setRenderer(new ManchesterOWLSyntaxOWLObjectRendererImpl());
- //set logging properties
- Logger.getLogger(SparqlKnowledgeSource.class).setLevel(Level.WARN);
- Logger.getLogger(ExtractionAlgorithm.class).setLevel(Level.WARN);
- Logger.getLogger(org.dllearner.kb.extraction.Manager.class).setLevel(Level.WARN);
- Logger.getRootLogger().removeAllAppenders();
- Logger.getRootLogger().addAppender(new ConsoleAppender(new PatternLayout("%m%n")));
- // KB2
- SparqlEndpoint endpoint1 = SparqlEndpoint.getEndpointDBpedia();
- ExtractionDBCache cache1 = new ExtractionDBCache("cache");
- String namespace1 = "http://dbpedia.org/resource/";
- KnowledgeBase kb1 = new KnowledgeBase(endpoint1, cache1, namespace1);
- // KB2
- SparqlEndpoint endpoint2 = new SparqlEndpoint(new URL("http://wifo5-03.informatik.uni-mannheim.de/factbook/sparql"));
- ExtractionDBCache cache2 = new ExtractionDBCache("cache");
- //TODO problem with World Factbook is that old FU Berlin server is useless because of bugs and current version
- //is provide by University Of Mannheim now with another namespace http://wifo5-03.informatik.uni-mannheim.de/factbook/resource/
- //but the DBpedia links are still to the old D2R server instance
- //workaround: replace namespace before learning
- String namespace2 = "http://www4.wiwiss.fu-berlin.de/factbook/resource/";
- KnowledgeBase kb2 = new KnowledgeBase(endpoint2, cache2, namespace2);
-
- OntologyMatching matcher = new OntologyMatching(kb1, kb2);
- matcher.start();
-
- }
-
}
Added: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java (rev 0)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatchingTest.java 2013-01-30 09:55:24 UTC (rev 3893)
@@ -0,0 +1,85 @@
+package org.dllearner.scripts;
+
+import java.net.URL;
+
+import org.apache.log4j.ConsoleAppender;
+import org.apache.log4j.Level;
+import org.apache.log4j.Logger;
+import org.apache.log4j.PatternLayout;
+import org.dllearner.kb.extraction.ExtractionAlgorithm;
+import org.dllearner.kb.sparql.ExtractionDBCache;
+import org.dllearner.kb.sparql.SparqlEndpoint;
+import org.dllearner.kb.sparql.SparqlKnowledgeSource;
+import org.dllearner.scripts.OntologyMatching.KnowledgeBase;
+import org.junit.Before;
+import org.junit.Test;
+import org.semanticweb.owlapi.io.ToStringRenderer;
+
+import uk.ac.manchester.cs.owl.owlapi.mansyntaxrenderer.ManchesterOWLSyntaxOWLObjectRendererImpl;
+
+public class OntologyMatchingTest {
+
+ private KnowledgeBase dbpedia;
+ private KnowledgeBase worldFactBook;
+ private KnowledgeBase openCyc;
+ private KnowledgeBase linkedGeoData;
+
+ @Before
+ public void setUp() throws Exception {
+ // render output
+ ToStringRenderer.getInstance().setRenderer(new ManchesterOWLSyntaxOWLObjectRendererImpl());
+ // set logging properties
+ Logger.getLogger(SparqlKnowledgeSource.class).setLevel(Level.WARN);
+ Logger.getLogger(ExtractionAlgorithm.class).setLevel(Level.WARN);
+ Logger.getLogger(org.dllearner.kb.extraction.Manager.class).setLevel(Level.WARN);
+ Logger.getRootLogger().removeAllAppenders();
+ Logger.getRootLogger().addAppender(new ConsoleAppender(new PatternLayout("%m%n")));
+
+ //DBpedia
+ SparqlEndpoint endpoint = SparqlEndpoint.getEndpointDBpedia();
+ ExtractionDBCache cache = new ExtractionDBCache("cache");
+ String namespace = "http://dbpedia.org/resource/";
+ dbpedia = new KnowledgeBase(endpoint, cache, namespace);
+
+ //World Factbook
+ //TODO problem with World Factbook is that old FU Berlin server is useless because of bugs and current version
+ //is provide by University Of Mannheim now with another namespace http://wifo5-03.informatik.uni-mannheim.de/factbook/resource/
+ //but the DBpedia links are still to the old D2R server instance
+ //workaround: replace namespace before learning
+ endpoint = new SparqlEndpoint(new URL("http://wifo5-03.informatik.uni-mannheim.de/factbook/sparql"));
+ cache = new ExtractionDBCache("cache");
+ namespace = "http://www4.wiwiss.fu-berlin.de/factbook/resource/";
+ worldFactBook = new KnowledgeBase(endpoint, cache, namespace);
+
+ //local OpenCyc
+ endpoint = new SparqlEndpoint(new URL("http://localhost:8890/sparql"));
+ cache = new ExtractionDBCache("cache");
+ namespace = "http://sw.cyc.com";
+ openCyc = new KnowledgeBase(endpoint, cache, namespace);
+
+ //LinkedGeoData
+ endpoint = new SparqlEndpoint(new URL("http://linkedgeodata.org/sparql"));
+ cache = new ExtractionDBCache("cache");
+ namespace = "http://linkedgeodata.org/triplify/";
+ linkedGeoData = new KnowledgeBase(endpoint, cache, namespace);
+ }
+
+ @Test
+ public void testDBpediaWorldFactbook() {
+ OntologyMatching matcher = new OntologyMatching(dbpedia, worldFactBook);
+ matcher.start();
+ }
+
+ @Test
+ public void testDBpediaOpenCyc() {
+ OntologyMatching matcher = new OntologyMatching(dbpedia, openCyc);
+ matcher.start();
+ }
+
+ @Test
+ public void testDBpediaLinkedGeoData() {
+ OntologyMatching matcher = new OntologyMatching(dbpedia, linkedGeoData);
+ matcher.start();
+ }
+
+}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-01-25 13:04:42
|
Revision: 3892
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3892&view=rev
Author: lorenz_b
Date: 2013-01-25 13:04:28 +0000 (Fri, 25 Jan 2013)
Log Message:
-----------
Continued matching script.
Modified Paths:
--------------
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-01-25 12:02:10 UTC (rev 3891)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-01-25 13:04:28 UTC (rev 3892)
@@ -1,19 +1,30 @@
package org.dllearner.scripts;
import java.net.URL;
+import java.text.DecimalFormat;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Map.Entry;
import java.util.Set;
import java.util.SortedSet;
import java.util.TreeSet;
+import org.apache.log4j.ConsoleAppender;
+import org.apache.log4j.Level;
import org.apache.log4j.Logger;
+import org.apache.log4j.PatternLayout;
import org.dllearner.algorithms.celoe.CELOE;
import org.dllearner.core.AbstractLearningProblem;
import org.dllearner.core.AbstractReasonerComponent;
import org.dllearner.core.ComponentInitException;
+import org.dllearner.core.EvaluatedDescription;
+import org.dllearner.core.owl.Description;
import org.dllearner.core.owl.Individual;
import org.dllearner.core.owl.NamedClass;
import org.dllearner.core.owl.ObjectProperty;
import org.dllearner.kb.SparqlEndpointKS;
+import org.dllearner.kb.extraction.ExtractionAlgorithm;
import org.dllearner.kb.sparql.ExtractionDBCache;
import org.dllearner.kb.sparql.SPARQLTasks;
import org.dllearner.kb.sparql.SparqlEndpoint;
@@ -26,7 +37,12 @@
import org.dllearner.utilities.datastructures.Datastructures;
import org.dllearner.utilities.datastructures.SortedSetTuple;
import org.dllearner.utilities.examples.AutomaticNegativeExampleFinderSPARQL2;
+import org.dllearner.utilities.owl.DLLearnerDescriptionConvertVisitor;
+import org.dllearner.utilities.owl.OWLAPIDescriptionConvertVisitor;
+import org.semanticweb.owlapi.io.ToStringRenderer;
+import uk.ac.manchester.cs.owl.owlapi.mansyntaxrenderer.ManchesterOWLSyntaxOWLObjectRendererImpl;
+
import com.hp.hpl.jena.query.Query;
import com.hp.hpl.jena.query.QueryFactory;
import com.hp.hpl.jena.query.QuerySolution;
@@ -63,11 +79,51 @@
}
public void start(){
- computeMatching(kb1, kb2);
- computeMatching(kb2, kb1);
+ Map<Description, List<? extends EvaluatedDescription>> mapping1 = computeMapping(kb1, kb2);
+ printMappingPretty(mapping1);
+ Map<Description, List<? extends EvaluatedDescription>> mapping2 = computeMapping(kb2, kb1);
+ printMappingPretty(mapping2);
}
- private void computeMatching(KnowledgeBase source, KnowledgeBase target) {
+ private void printMapping(Map<Description, List<? extends EvaluatedDescription>> mapping){
+ logger.info("Source Class -> Target Class Expression");
+ for (Entry<Description, List<? extends org.dllearner.core.EvaluatedDescription>> entry : mapping.entrySet()) {
+ Description key = entry.getKey();
+ int length = key.toString().length();
+ String indention = "";
+ for(int i = 0; i < length; i++){
+ indention += " ";
+ }
+ List<? extends org.dllearner.core.EvaluatedDescription> value = entry.getValue();
+ logger.info(key.toString());
+ for (EvaluatedDescription evaluatedDescription : value) {
+ logger.info(indention + "\t->\t" + evaluatedDescription);
+ }
+ }
+ }
+
+ private void printMappingPretty(Map<Description, List<? extends EvaluatedDescription>> mapping){
+ DecimalFormat dfPercent = new DecimalFormat("0.00%");
+ logger.info("Source Class -> Target Class Expression");
+ for (Entry<Description, List<? extends org.dllearner.core.EvaluatedDescription>> entry : mapping.entrySet()) {
+ Description key = entry.getKey();
+ int length = OWLAPIDescriptionConvertVisitor.getOWLClassExpression(key).toString().length();
+ String indention = "";
+ for(int i = 0; i < length; i++){
+ indention += " ";
+ }
+ List<? extends org.dllearner.core.EvaluatedDescription> value = entry.getValue();
+ logger.info(OWLAPIDescriptionConvertVisitor.getOWLClassExpression(key));
+ for (EvaluatedDescription evaluatedDescription : value) {
+ logger.info(indention + "\t->\t" +
+ OWLAPIDescriptionConvertVisitor.getOWLClassExpression(evaluatedDescription.getDescription()) +
+ "(" + dfPercent.format(evaluatedDescription.getAccuracy()) + ")");
+ }
+ }
+ }
+
+ private Map<Description, List<? extends EvaluatedDescription>> computeMapping(KnowledgeBase source, KnowledgeBase target) {
+ Map<Description, List<? extends EvaluatedDescription>> mapping = new HashMap<Description, List<? extends EvaluatedDescription>>();
// get all classes in SOURCE
Set<NamedClass> sourceClasses = getClasses(source);
@@ -79,16 +135,18 @@
logger.info(individuals);
//learn concept in KB2 based on the examples
if(individuals.size() >= 3){
- learnClassExpression(target, individuals);
+ List<? extends EvaluatedDescription> learnedClassExpressions = learnClassExpression(target, individuals);
+ mapping.put(nc, learnedClassExpressions);
}
}
+ return mapping;
}
- private void learnClassExpression(KnowledgeBase kb, SortedSet<Individual> posExamples){
- learnClassExpression(kb, posExamples, false);
+ private List<? extends EvaluatedDescription> learnClassExpression(KnowledgeBase kb, SortedSet<Individual> posExamples){
+ return learnClassExpression(kb, posExamples, false);
}
- private void learnClassExpression(KnowledgeBase kb, SortedSet<Individual> positiveExamples, boolean posNeg){
+ private List<? extends EvaluatedDescription> learnClassExpression(KnowledgeBase kb, SortedSet<Individual> positiveExamples, boolean posNeg){
try {
SortedSet<Individual> negativeExamples = new TreeSet<Individual>();
if(posNeg){
@@ -132,11 +190,14 @@
la.setNoisePercentage(25);
la.init();
la.start();
+
+ logger.info(la.getCurrentlyBestEvaluatedDescription());
- logger.info(la.getCurrentlyBestEvaluatedDescription());
+ return la.getCurrentlyBestEvaluatedDescriptions(10);
} catch (ComponentInitException e) {
e.printStackTrace();
}
+ return null;
}
private Set<NamedClass> getClasses(KnowledgeBase kb){
@@ -195,7 +256,11 @@
qs = rs.next();
RDFNode object = qs.get("o");
if(object.isURIResource()){
- relatedIndividuals.add(new Individual(object.asResource().getURI()));
+
+ String uri = object.asResource().getURI();
+ //workaround for World Factbook - should be removed later
+ uri = uri.replace("http://www4.wiwiss.fu-berlin.de/factbook/resource/", "http://wifo5-03.informatik.uni-mannheim.de/factbook/resource/");
+ relatedIndividuals.add(new Individual(uri));
}
}
return relatedIndividuals;
@@ -272,6 +337,14 @@
* @param args
*/
public static void main(String[] args) throws Exception{
+ //render output
+ ToStringRenderer.getInstance().setRenderer(new ManchesterOWLSyntaxOWLObjectRendererImpl());
+ //set logging properties
+ Logger.getLogger(SparqlKnowledgeSource.class).setLevel(Level.WARN);
+ Logger.getLogger(ExtractionAlgorithm.class).setLevel(Level.WARN);
+ Logger.getLogger(org.dllearner.kb.extraction.Manager.class).setLevel(Level.WARN);
+ Logger.getRootLogger().removeAllAppenders();
+ Logger.getRootLogger().addAppender(new ConsoleAppender(new PatternLayout("%m%n")));
// KB2
SparqlEndpoint endpoint1 = SparqlEndpoint.getEndpointDBpedia();
ExtractionDBCache cache1 = new ExtractionDBCache("cache");
@@ -280,6 +353,10 @@
// KB2
SparqlEndpoint endpoint2 = new SparqlEndpoint(new URL("http://wifo5-03.informatik.uni-mannheim.de/factbook/sparql"));
ExtractionDBCache cache2 = new ExtractionDBCache("cache");
+ //TODO problem with World Factbook is that old FU Berlin server is useless because of bugs and current version
+ //is provide by University Of Mannheim now with another namespace http://wifo5-03.informatik.uni-mannheim.de/factbook/resource/
+ //but the DBpedia links are still to the old D2R server instance
+ //workaround: replace namespace before learning
String namespace2 = "http://www4.wiwiss.fu-berlin.de/factbook/resource/";
KnowledgeBase kb2 = new KnowledgeBase(endpoint2, cache2, namespace2);
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-01-25 12:02:22
|
Revision: 3891
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3891&view=rev
Author: lorenz_b
Date: 2013-01-25 12:02:10 +0000 (Fri, 25 Jan 2013)
Log Message:
-----------
Started script for instance based matching of 2 SPARQL knowledge bases.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/kb/sparql/SPARQLTasks.java
trunk/components-core/src/main/java/org/dllearner/learningproblems/PosOnlyLP.java
trunk/scripts/pom.xml
trunk/scripts/src/main/java/org/dllearner/scripts/matching/GeneralMatcher.java
trunk/test/qtl/dbpedia_simple.conf
Added Paths:
-----------
trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
Modified: trunk/components-core/src/main/java/org/dllearner/kb/sparql/SPARQLTasks.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/kb/sparql/SPARQLTasks.java 2013-01-23 13:30:14 UTC (rev 3890)
+++ trunk/components-core/src/main/java/org/dllearner/kb/sparql/SPARQLTasks.java 2013-01-25 12:02:10 UTC (rev 3891)
@@ -701,7 +701,7 @@
public Set<NamedClass> getAllClasses() {
Set<NamedClass> classes = new TreeSet<NamedClass>();
- String query = "PREFIX owl: <http://www.w3.org/2002/07/owl#> SELECT ?c WHERE {?c a owl:Class} LIMIT 1000";
+ String query = "SELECT ?c WHERE {?c a <http://www.w3.org/2002/07/owl#Class>} LIMIT 1000";
/*
* String query = "PREFIX owl: <http://www.w3.org/2002/07/owl#> PREFIX rdfs:<http://www.w3.org/2000/01/rdf-schema#> " +
"SELECT ?c WHERE {{?c a owl:Class} UNION {?c rdfs:subClassOf ?d} UNION {?d rdfs:subClassOf ?c}} LIMIT 1000";
Modified: trunk/components-core/src/main/java/org/dllearner/learningproblems/PosOnlyLP.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/learningproblems/PosOnlyLP.java 2013-01-23 13:30:14 UTC (rev 3890)
+++ trunk/components-core/src/main/java/org/dllearner/learningproblems/PosOnlyLP.java 2013-01-25 12:02:10 UTC (rev 3891)
@@ -64,6 +64,11 @@
public PosOnlyLP(AbstractReasonerComponent reasoningService) {
super(reasoningService);
}
+
+ public PosOnlyLP(AbstractReasonerComponent reasoningService, SortedSet<Individual> positiveExamples) {
+ super(reasoningService);
+ this.positiveExamples = positiveExamples;
+ }
/*
* (non-Javadoc)
Modified: trunk/scripts/pom.xml
===================================================================
--- trunk/scripts/pom.xml 2013-01-23 13:30:14 UTC (rev 3890)
+++ trunk/scripts/pom.xml 2013-01-25 12:02:10 UTC (rev 3891)
@@ -25,6 +25,10 @@
<artifactId>jena</artifactId>
<groupId>com.hp.hpl.jena</groupId>
</exclusion>
+ <exclusion>
+ <artifactId>owlapi</artifactId>
+ <groupId>net.sourceforge.owlapi</groupId>
+ </exclusion>
</exclusions>
</dependency>
<dependency>
@@ -40,6 +44,12 @@
<dependency>
<groupId>org.dllearner</groupId>
<artifactId>interfaces</artifactId>
+ <exclusions>
+ <exclusion>
+ <artifactId>owlapi</artifactId>
+ <groupId>net.sourceforge.owlapi</groupId>
+ </exclusion>
+ </exclusions>
</dependency>
<dependency>
<groupId>net.sourceforge.secondstring</groupId>
@@ -49,12 +59,12 @@
<groupId>postgresql</groupId>
<artifactId>postgresql</artifactId>
</dependency>
- <dependency>
- <groupId>com.dumontierlab</groupId>
- <artifactId>pdb2rdf-parser</artifactId>
- </dependency>
<dependency>
<groupId>com.dumontierlab</groupId>
+ <artifactId>pdb2rdf-parser</artifactId>
+ </dependency>
+ <dependency>
+ <groupId>com.dumontierlab</groupId>
<artifactId>pdb2rdf-cli</artifactId>
<exclusions>
<exclusion>
@@ -84,8 +94,8 @@
<groupId>org.deri.any23</groupId>
</exclusion>
<exclusion>
- <artifactId>jena-core</artifactId>
- <groupId>org.apache.jena</groupId>
+ <artifactId>jena-core</artifactId>
+ <groupId>org.apache.jena</groupId>
</exclusion>
</exclusions>
</dependency>
@@ -103,6 +113,30 @@
<artifactId>weka</artifactId>
<version>3.6.5</version>
</dependency>
+ <dependency>
+ <groupId>net.sourceforge.owlapi</groupId>
+ <artifactId>owlapi-distribution</artifactId>
+ </dependency>
+ <dependency>
+ <groupId>net.sourceforge.owlapi</groupId>
+ <artifactId>owlapi-reasoner</artifactId>
+ <exclusions>
+ <exclusion>
+ <artifactId>owlapi-api</artifactId>
+ <groupId>net.sourceforge.owlapi</groupId>
+ </exclusion>
+ </exclusions>
+ </dependency>
+ <dependency>
+ <groupId>net.sourceforge.owlapi</groupId>
+ <artifactId>owlapi-util</artifactId>
+ <exclusions>
+ <exclusion>
+ <artifactId>owlapi-api</artifactId>
+ <groupId>net.sourceforge.owlapi</groupId>
+ </exclusion>
+ </exclusions>
+ </dependency>
</dependencies>
Added: trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java (rev 0)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/OntologyMatching.java 2013-01-25 12:02:10 UTC (rev 3891)
@@ -0,0 +1,291 @@
+package org.dllearner.scripts;
+
+import java.net.URL;
+import java.util.Set;
+import java.util.SortedSet;
+import java.util.TreeSet;
+
+import org.apache.log4j.Logger;
+import org.dllearner.algorithms.celoe.CELOE;
+import org.dllearner.core.AbstractLearningProblem;
+import org.dllearner.core.AbstractReasonerComponent;
+import org.dllearner.core.ComponentInitException;
+import org.dllearner.core.owl.Individual;
+import org.dllearner.core.owl.NamedClass;
+import org.dllearner.core.owl.ObjectProperty;
+import org.dllearner.kb.SparqlEndpointKS;
+import org.dllearner.kb.sparql.ExtractionDBCache;
+import org.dllearner.kb.sparql.SPARQLTasks;
+import org.dllearner.kb.sparql.SparqlEndpoint;
+import org.dllearner.kb.sparql.SparqlKnowledgeSource;
+import org.dllearner.kb.sparql.SparqlQuery;
+import org.dllearner.learningproblems.PosNegLPStandard;
+import org.dllearner.learningproblems.PosOnlyLP;
+import org.dllearner.reasoning.FastInstanceChecker;
+import org.dllearner.reasoning.SPARQLReasoner;
+import org.dllearner.utilities.datastructures.Datastructures;
+import org.dllearner.utilities.datastructures.SortedSetTuple;
+import org.dllearner.utilities.examples.AutomaticNegativeExampleFinderSPARQL2;
+
+import com.hp.hpl.jena.query.Query;
+import com.hp.hpl.jena.query.QueryFactory;
+import com.hp.hpl.jena.query.QuerySolution;
+import com.hp.hpl.jena.query.ResultSet;
+import com.hp.hpl.jena.query.Syntax;
+import com.hp.hpl.jena.rdf.model.RDFNode;
+import com.hp.hpl.jena.sparql.engine.http.QueryEngineHTTP;
+import com.hp.hpl.jena.vocabulary.OWL;
+import com.jamonapi.Monitor;
+import com.jamonapi.MonitorFactory;
+
+public class OntologyMatching {
+
+
+ private static final Logger logger = Logger.getLogger(OntologyMatching.class.getName());
+
+ private final ObjectProperty sameAs = new ObjectProperty(OWL.sameAs.getURI());
+ private final Monitor mon;
+
+ //KB1
+ private KnowledgeBase kb1;
+ //KB2
+ private KnowledgeBase kb2;
+
+ public OntologyMatching(KnowledgeBase kb1, KnowledgeBase kb2) {
+ this.kb1 = kb1;
+ this.kb2 = kb2;
+
+ mon = MonitorFactory.getTimeMonitor("time");
+ }
+
+ public OntologyMatching(SparqlEndpoint endpoint1, SparqlEndpoint endpoint2) {
+ this(new KnowledgeBase(endpoint1), new KnowledgeBase(endpoint2));
+ }
+
+ public void start(){
+ computeMatching(kb1, kb2);
+ computeMatching(kb2, kb1);
+ }
+
+ private void computeMatching(KnowledgeBase source, KnowledgeBase target) {
+ // get all classes in SOURCE
+ Set<NamedClass> sourceClasses = getClasses(source);
+
+ // for each class of KB1
+ for (NamedClass nc : sourceClasses) {
+ logger.info(nc);
+ // get all via owl:sameAs related individuals
+ SortedSet<Individual> individuals = getRelatedIndividualsNamespaceAware(source, nc, target.getNamespace());
+ logger.info(individuals);
+ //learn concept in KB2 based on the examples
+ if(individuals.size() >= 3){
+ learnClassExpression(target, individuals);
+ }
+ }
+ }
+
+ private void learnClassExpression(KnowledgeBase kb, SortedSet<Individual> posExamples){
+ learnClassExpression(kb, posExamples, false);
+ }
+
+ private void learnClassExpression(KnowledgeBase kb, SortedSet<Individual> positiveExamples, boolean posNeg){
+ try {
+ SortedSet<Individual> negativeExamples = new TreeSet<Individual>();
+ if(posNeg){
+ //find negative examples
+ mon.start();
+ AutomaticNegativeExampleFinderSPARQL2 finder = new AutomaticNegativeExampleFinderSPARQL2(kb.getEndpoint());
+ //TODO find negative examples
+ mon.stop();
+ logger.info("Found " + negativeExamples.size() + " negative examples in " + mon.getLastValue() + "ms.");
+ }
+
+ SortedSetTuple<Individual> examples = new SortedSetTuple<Individual>(positiveExamples, negativeExamples);
+
+ SparqlKnowledgeSource ks = new SparqlKnowledgeSource();
+ ks.setInstances(Datastructures.individualSetToStringSet(examples.getCompleteSet()));
+ ks.setUrl(kb.getEndpoint().getURL());
+ ks.setDefaultGraphURIs(new TreeSet<String>(kb.getEndpoint().getDefaultGraphURIs()));
+ ks.setUseLits(false);
+ ks.setUseCacheDatabase(true);
+ ks.setCacheDir("cache");
+ ks.setRecursionDepth(2);
+ ks.setCloseAfterRecursion(true);
+ ks.setDissolveBlankNodes(false);
+ ks.setSaveExtractedFragment(false);
+ ks.init();
+
+ AbstractReasonerComponent rc = new FastInstanceChecker(ks);
+ rc.init();
+
+ AbstractLearningProblem lp;
+ if(posNeg){
+ lp = new PosNegLPStandard(rc, positiveExamples, negativeExamples);
+ } else {
+ lp = new PosOnlyLP(rc, positiveExamples);
+
+ }
+ lp.init();
+
+ CELOE la = new CELOE(lp, rc);
+ la.setMaxExecutionTimeInSeconds(10);
+ la.setNoisePercentage(25);
+ la.init();
+ la.start();
+
+ logger.info(la.getCurrentlyBestEvaluatedDescription());
+ } catch (ComponentInitException e) {
+ e.printStackTrace();
+ }
+ }
+
+ private Set<NamedClass> getClasses(KnowledgeBase kb){
+ Set<NamedClass> classes = kb.getSparqlHelper().getAllClasses();
+ //fallback: check for ?s a ?type where ?type is not asserted to owl:Class
+ if(classes.isEmpty()){
+ String query = "SELECT ?type WHERE {?s a ?type.}";
+ ResultSet rs = executeSelect(kb, query);
+ QuerySolution qs;
+ while(rs.hasNext()){
+ qs = rs.next();
+ if(qs.get("type").isURIResource()){
+ classes.add(new NamedClass(qs.get("type").asResource().getURI()));
+ }
+ }
+ }
+ return classes;
+ }
+
+ private SortedSet<Individual> getRelatedIndividualsNaive(KnowledgeBase kb, NamedClass nc){
+ SortedSet<Individual> relatedIndividuals = new TreeSet<Individual>();
+ //get all individuals in given class nc
+ Set<Individual> individuals = kb.getReasoner().getIndividuals(nc);
+ //for each individual in class nc
+ for(Individual ind : individuals){
+ //get all individuals related via owl:sameAs
+ Set<Individual> sameIndividuals = kb.getReasoner().getRelatedIndividuals(ind, sameAs);
+ relatedIndividuals.addAll(sameIndividuals);
+ }
+ return relatedIndividuals;
+ }
+
+ private SortedSet<Individual> getRelatedIndividuals(KnowledgeBase kb, NamedClass nc){
+ SortedSet<Individual> relatedIndividuals = new TreeSet<Individual>();
+ //get all individuals o which are connected to individuals s belonging to class nc
+ String query = String.format("SELECT ?o WHERE {?s a <%s>. ?s <http://www.w3.org/2002/07/owl#sameAs> ?o.}", nc.getName());
+ ResultSet rs = executeSelect(kb, query);
+ QuerySolution qs;
+ while(rs.hasNext()){
+ qs = rs.next();
+ RDFNode object = qs.get("o");
+ if(object.isURIResource()){
+ relatedIndividuals.add(new Individual(object.asResource().getURI()));
+ }
+ }
+ return relatedIndividuals;
+ }
+
+ private SortedSet<Individual> getRelatedIndividualsNamespaceAware(KnowledgeBase kb, NamedClass nc, String targetNamespace){
+ SortedSet<Individual> relatedIndividuals = new TreeSet<Individual>();
+ //get all individuals o which are connected to individuals s belonging to class nc
+ String query = String.format("SELECT ?o WHERE {?s a <%s>. ?s <http://www.w3.org/2002/07/owl#sameAs> ?o. FILTER(REGEX(STR(?o),'%s'))}", nc.getName(), targetNamespace);
+ ResultSet rs = executeSelect(kb, query);
+ QuerySolution qs;
+ while(rs.hasNext()){
+ qs = rs.next();
+ RDFNode object = qs.get("o");
+ if(object.isURIResource()){
+ relatedIndividuals.add(new Individual(object.asResource().getURI()));
+ }
+ }
+ return relatedIndividuals;
+ }
+
+ protected ResultSet executeSelect(KnowledgeBase kb, String query){
+ return executeSelect(kb, QueryFactory.create(query, Syntax.syntaxARQ));
+ }
+
+ protected ResultSet executeSelect(KnowledgeBase kb, Query query){
+ ExtractionDBCache cache = kb.getCache();
+ SparqlEndpoint endpoint = kb.getEndpoint();
+ ResultSet rs = null;
+ if(cache != null){
+ rs = SparqlQuery.convertJSONtoResultSet(cache.executeSelectQuery(endpoint, query.toString()));
+ } else {
+ QueryEngineHTTP qe = new QueryEngineHTTP(endpoint.getURL().toString(), query);
+ for(String uri : endpoint.getDefaultGraphURIs()){
+ qe.addDefaultGraph(uri);
+ }
+ rs = qe.execSelect();
+ }
+ return rs;
+ }
+
+ public static class KnowledgeBase{
+ private SparqlEndpoint endpoint;
+ private SPARQLReasoner reasoner;
+ private SPARQLTasks sparqlHelper;
+ private String namespace;
+ private ExtractionDBCache cache;
+
+ public KnowledgeBase(SparqlEndpoint endpoint, ExtractionDBCache cache, String namespace) {
+ this.endpoint = endpoint;
+ this.namespace = namespace;
+ this.cache = cache;
+
+ this.reasoner = new SPARQLReasoner(new SparqlEndpointKS(endpoint));
+ this.sparqlHelper = new SPARQLTasks(endpoint);
+ }
+
+ public KnowledgeBase(SparqlEndpoint endpoint) {
+ this.endpoint = endpoint;
+
+ this.reasoner = new SPARQLReasoner(new SparqlEndpointKS(endpoint));
+ this.sparqlHelper = new SPARQLTasks(endpoint);
+ }
+
+ public SparqlEndpoint getEndpoint() {
+ return endpoint;
+ }
+
+ public SPARQLReasoner getReasoner() {
+ return reasoner;
+ }
+
+ public SPARQLTasks getSparqlHelper() {
+ return sparqlHelper;
+ }
+
+ public String getNamespace() {
+ return namespace;
+ }
+
+ public ExtractionDBCache getCache() {
+ return cache;
+ }
+
+
+
+ }
+
+ /**
+ * @param args
+ */
+ public static void main(String[] args) throws Exception{
+ // KB2
+ SparqlEndpoint endpoint1 = SparqlEndpoint.getEndpointDBpedia();
+ ExtractionDBCache cache1 = new ExtractionDBCache("cache");
+ String namespace1 = "http://dbpedia.org/resource/";
+ KnowledgeBase kb1 = new KnowledgeBase(endpoint1, cache1, namespace1);
+ // KB2
+ SparqlEndpoint endpoint2 = new SparqlEndpoint(new URL("http://wifo5-03.informatik.uni-mannheim.de/factbook/sparql"));
+ ExtractionDBCache cache2 = new ExtractionDBCache("cache");
+ String namespace2 = "http://www4.wiwiss.fu-berlin.de/factbook/resource/";
+ KnowledgeBase kb2 = new KnowledgeBase(endpoint2, cache2, namespace2);
+
+ OntologyMatching matcher = new OntologyMatching(kb1, kb2);
+ matcher.start();
+
+ }
+
+}
Modified: trunk/scripts/src/main/java/org/dllearner/scripts/matching/GeneralMatcher.java
===================================================================
--- trunk/scripts/src/main/java/org/dllearner/scripts/matching/GeneralMatcher.java 2013-01-23 13:30:14 UTC (rev 3890)
+++ trunk/scripts/src/main/java/org/dllearner/scripts/matching/GeneralMatcher.java 2013-01-25 12:02:10 UTC (rev 3891)
@@ -24,7 +24,7 @@
import java.util.LinkedList;
import java.util.Set;
-import org.dllearner.algorithm.qtl.QTL;
+import org.dllearner.algorithms.qtl.QTL;
import org.dllearner.core.ComponentInitException;
import org.dllearner.core.LearningProblemUnsupportedException;
import org.dllearner.kb.SparqlEndpointKS;
Modified: trunk/test/qtl/dbpedia_simple.conf
===================================================================
--- trunk/test/qtl/dbpedia_simple.conf 2013-01-23 13:30:14 UTC (rev 3890)
+++ trunk/test/qtl/dbpedia_simple.conf 2013-01-25 12:02:10 UTC (rev 3891)
@@ -2,7 +2,7 @@
prefixes = [ ("dbr","http://dbpedia.org/resource/") ]
ks.type = "SPARQL"
-ks.url = "http://live.dbpedia.org/sparql"
+ks.url = "http://dbpedia.org/sparql"
ks.defaultGraphURIs = { "http://dbpedia.org" }
// learning problem
@@ -10,3 +10,4 @@
lp.positiveExamples = { "dbr:Bob_Dylan", "dbr:The_Beatles", "dbr:Aerosmith" }
alg.type = "qtl"
+alg.maxQueryTreeDepth = 1
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-01-23 13:30:23
|
Revision: 3890
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3890&view=rev
Author: lorenz_b
Date: 2013-01-23 13:30:14 +0000 (Wed, 23 Jan 2013)
Log Message:
-----------
Added option to enable simple RDFS reasoning for local model based SPARQL endpoints.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java
Modified: trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java 2013-01-22 21:41:16 UTC (rev 3889)
+++ trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java 2013-01-23 13:30:14 UTC (rev 3890)
@@ -19,7 +19,8 @@
private OntModel model;
private String fileName;
- private String baseDir;
+ private String baseDir;
+ private boolean enableReasoning = false;
public LocalModelBasedSparqlEndpointKS() {
}
@@ -28,22 +29,6 @@
this(new URL(ontologyURL));
}
- public void setFileName(String fileName) {
- this.fileName = fileName;
- }
-
- public String getFileName() {
- return fileName;
- }
-
- public String getBaseDir() {
- return baseDir;
- }
-
- public void setBaseDir(String baseDir) {
- this.baseDir = baseDir;
- }
-
public LocalModelBasedSparqlEndpointKS(URL ontologyURL) {
this.fileName = ontologyURL.toString();
}
@@ -54,7 +39,7 @@
@Override
public void init() throws ComponentInitException {
- Model baseModel = ModelFactory.createDefaultModel();System.out.println("Base: " + baseDir);
+ Model baseModel = ModelFactory.createDefaultModel();
// use the FileManager to find the input file
InputStream in = FileManager.get().open(baseDir + File.separator + fileName);
if (in == null) {
@@ -64,9 +49,33 @@
// read the RDF/XML file
baseModel.read(in, null);
- model = ModelFactory.createOntologyModel(OntModelSpec.RDFS_MEM_RDFS_INF, baseModel);
+ model = ModelFactory.createOntologyModel(enableReasoning ? OntModelSpec.OWL_MEM : OntModelSpec.OWL_MEM_RDFS_INF, baseModel);
}
+ public void setFileName(String fileName) {
+ this.fileName = fileName;
+ }
+
+ public String getFileName() {
+ return fileName;
+ }
+
+ public String getBaseDir() {
+ return baseDir;
+ }
+
+ public void setBaseDir(String baseDir) {
+ this.baseDir = baseDir;
+ }
+
+ public void setEnableReasoning(boolean enableReasoning) {
+ this.enableReasoning = enableReasoning;
+ }
+
+ public boolean isEnableReasoning() {
+ return enableReasoning;
+ }
+
public OntModel getModel() {
return model;
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-01-22 21:41:24
|
Revision: 3889
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3889&view=rev
Author: lorenz_b
Date: 2013-01-22 21:41:16 +0000 (Tue, 22 Jan 2013)
Log Message:
-----------
Added options to set prefixes in QTL output and to enable numeric SPARQL FILTER expression learning.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java
trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/datastructures/QueryTree.java
trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/datastructures/impl/QueryTreeImpl.java
trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/impl/QueryTreeFactoryImpl.java
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java 2013-01-22 13:22:29 UTC (rev 3888)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java 2013-01-22 21:41:16 UTC (rev 3889)
@@ -22,9 +22,11 @@
import java.util.ArrayList;
import java.util.Collection;
import java.util.Collections;
+import java.util.HashMap;
import java.util.HashSet;
import java.util.LinkedList;
import java.util.List;
+import java.util.Map;
import java.util.Set;
import java.util.SortedSet;
import java.util.TreeSet;
@@ -116,6 +118,8 @@
private SortedSet<String> lggInstances;
private Set<String> objectNamespacesToIgnore = new HashSet<String>();
+ private Map<String, String> prefixes = new HashMap<String, String>();
+ private boolean enableNumericLiteralFilters = false;
public static Collection<ConfigOption<?>> createConfigOptions() {
Collection<ConfigOption<?>> options = new LinkedList<ConfigOption<?>>();
@@ -125,7 +129,6 @@
}
public QTL() {
-
}
public QTL(AbstractLearningProblem learningProblem, SparqlEndpointKS endpointKS) throws LearningProblemUnsupportedException{
@@ -219,6 +222,14 @@
return maxQueryTreeDepth;
}
+ public void setPrefixes(Map<String, String> prefixes) {
+ this.prefixes = prefixes;
+ }
+
+ public Map<String, String> getPrefixes() {
+ return prefixes;
+ }
+
public String getSPARQLQuery(){
if(lgg == null){
lgg = lggGenerator.getLGG(getQueryTrees(posExamples));
@@ -256,6 +267,7 @@
if(logger.isDebugEnabled()){
logger.debug("Tree for resource " + resource);
logger.debug(tree.getStringRepresentation());
+
}
trees.add(tree);
}
@@ -330,9 +342,17 @@
if(logger.isDebugEnabled()){
logger.debug("LGG: \n" + lgg.getStringRepresentation());
}
- logger.info(lgg.toSPARQLQuery());
+ logger.info(lgg.toSPARQLQueryString(enableNumericLiteralFilters, prefixes));
}
-
+
+ public void setEnableNumericLiteralFilters(boolean enableNumericLiteralFilters) {
+ this.enableNumericLiteralFilters = enableNumericLiteralFilters;
+ }
+
+ public boolean isEnableNumericLiteralFilters() {
+ return enableNumericLiteralFilters;
+ }
+
@Override
public List<String> getCurrentlyBestSPARQLQueries(int nrOfSPARQLQueries) {
return Collections.singletonList(getBestSPARQLQuery());
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/datastructures/QueryTree.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/datastructures/QueryTree.java 2013-01-22 13:22:29 UTC (rev 3888)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/datastructures/QueryTree.java 2013-01-22 21:41:16 UTC (rev 3889)
@@ -22,6 +22,7 @@
import java.io.PrintWriter;
import java.util.Comparator;
import java.util.List;
+import java.util.Map;
import java.util.Set;
import org.dllearner.algorithms.qtl.datastructures.impl.QueryTreeImpl;
@@ -131,6 +132,8 @@
String toSPARQLQueryString(boolean filtered);
+ String toSPARQLQueryString(boolean filtered, Map<String, String> prefixMap);
+
Query toSPARQLQuery();
int getTriplePatternCount();
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/datastructures/impl/QueryTreeImpl.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/datastructures/impl/QueryTreeImpl.java 2013-01-22 13:22:29 UTC (rev 3888)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/datastructures/impl/QueryTreeImpl.java 2013-01-22 21:41:16 UTC (rev 3889)
@@ -31,6 +31,7 @@
import java.util.LinkedList;
import java.util.List;
import java.util.Map;
+import java.util.Map.Entry;
import java.util.Set;
import java.util.TreeSet;
import java.util.regex.Pattern;
@@ -712,6 +713,11 @@
@Override
public String toSPARQLQueryString(boolean filtered) {
+ return toSPARQLQueryString(filtered, Collections.<String, String>emptyMap());
+ }
+
+ @Override
+ public String toSPARQLQueryString(boolean filtered, Map<String, String> prefixMap) {
if(children.isEmpty()){
return "SELECT ?x0 WHERE {?x0 ?y ?z.}";
}
@@ -724,7 +730,16 @@
sb.append(filter).append("\n");
}
sb.append("}");
- return sb.toString();
+ Query query = QueryFactory.create(sb.toString(), Syntax.syntaxARQ);
+ query.setPrefix("rdf", "http://www.w3.org/1999/02/22-rdf-syntax-ns#");
+ query.setPrefix("rdfs", "http://www.w3.org/2000/01/rdf-schema#");
+
+ for (Entry<String, String> entry : prefixMap.entrySet()) {
+ String key = entry.getKey();
+ String value = entry.getValue();
+ query.setPrefix(key, value);
+ }
+ return query.toString();
}
private void buildSPARQLQueryString(QueryTree<N> tree, StringBuilder sb, boolean filtered, List<String> filters){
@@ -785,7 +800,7 @@
l = iter.next();
if(l.getDatatype() == XSDDatatype.XSDinteger || l.getDatatype() == XSDDatatype.XSDint){
min = (l.getInt() < min.getInt()) ? l : min;
- } else if(l.getDatatype() == XSDDatatype.XSDdouble){
+ } else if(l.getDatatype() == XSDDatatype.XSDdouble || l.getDatatype() == XSDDatatype.XSDdecimal){
min = (l.getDouble() < min.getDouble()) ? l : min;
} else if(l.getDatatype() == XSDDatatype.XSDdate){
min = (DatatypeConverter.parseDate(l.getLexicalForm()).compareTo(DatatypeConverter.parseDate(min.getLexicalForm())) == -1) ? l : min;
@@ -802,7 +817,7 @@
l = iter.next();
if(l.getDatatype() == XSDDatatype.XSDinteger || l.getDatatype() == XSDDatatype.XSDint){
max = (l.getInt() > max.getInt()) ? l : max;
- } else if(l.getDatatype() == XSDDatatype.XSDdouble){
+ } else if(l.getDatatype() == XSDDatatype.XSDdouble || l.getDatatype() == XSDDatatype.XSDdecimal){
max = (l.getDouble() > max.getDouble()) ? l : max;
} else if(l.getDatatype() == XSDDatatype.XSDdate){
max = (DatatypeConverter.parseDate(l.getLexicalForm()).compareTo(DatatypeConverter.parseDate(max.getLexicalForm())) == 1) ? l : max;
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/impl/QueryTreeFactoryImpl.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/impl/QueryTreeFactoryImpl.java 2013-01-22 13:22:29 UTC (rev 3888)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/impl/QueryTreeFactoryImpl.java 2013-01-22 21:41:16 UTC (rev 3889)
@@ -247,7 +247,8 @@
if(lit.getDatatype() == XSDDatatype.XSDinteger
|| lit.getDatatype() == XSDDatatype.XSDdouble
|| lit.getDatatype() == XSDDatatype.XSDdate
- || lit.getDatatype() == XSDDatatype.XSDint){
+ || lit.getDatatype() == XSDDatatype.XSDint
+ || lit.getDatatype() == XSDDatatype.XSDdecimal){
subTree.addLiteral(lit);
}
tree.addChild(subTree, st.getPredicate().toString());
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
Revision: 3888
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3888&view=rev
Author: lorenz_b
Date: 2013-01-22 13:22:29 +0000 (Tue, 22 Jan 2013)
Log Message:
-----------
Enabled config option to apply QTL on a local file with relativ path.
Modified Paths:
--------------
trunk/interfaces/src/main/java/org/dllearner/configuration/spring/ConfigurationBasedBeanDefinitionRegistryPostProcessor.java
Modified: trunk/interfaces/src/main/java/org/dllearner/configuration/spring/ConfigurationBasedBeanDefinitionRegistryPostProcessor.java
===================================================================
--- trunk/interfaces/src/main/java/org/dllearner/configuration/spring/ConfigurationBasedBeanDefinitionRegistryPostProcessor.java 2013-01-22 13:21:59 UTC (rev 3887)
+++ trunk/interfaces/src/main/java/org/dllearner/configuration/spring/ConfigurationBasedBeanDefinitionRegistryPostProcessor.java 2013-01-22 13:22:29 UTC (rev 3888)
@@ -3,6 +3,7 @@
import org.dllearner.configuration.IConfiguration;
import org.dllearner.configuration.IConfigurationProperty;
import org.dllearner.kb.KBFile;
+import org.dllearner.kb.LocalModelBasedSparqlEndpointKS;
import org.dllearner.kb.OWLFile;
import org.springframework.beans.BeansException;
import org.springframework.beans.factory.config.BeanDefinition;
@@ -95,7 +96,7 @@
/**
* Add Base Directory Value to Beans which need it.
*
- * @param beanDefinition The curren Bean Definition
+ * @param beanDefinition The current Bean Definition
*/
private void addBaseDirectoryIfNeeded(BeanDefinition beanDefinition) {
Class beanClass = null;
@@ -105,7 +106,7 @@
throw new RuntimeException("Can't find class " + beanDefinition.getBeanClassName());
}
/** Add Base Directory */
- if (beanClass.isAssignableFrom(KBFile.class) || beanClass.isAssignableFrom(OWLFile.class)) {
+ if (beanClass.isAssignableFrom(KBFile.class) || beanClass.isAssignableFrom(OWLFile.class) || beanClass.isAssignableFrom(LocalModelBasedSparqlEndpointKS.class)) {
beanDefinition.getPropertyValues().add("baseDir", configuration.getBaseDir());
}
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-01-22 13:22:07
|
Revision: 3887
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3887&view=rev
Author: lorenz_b
Date: 2013-01-22 13:21:59 +0000 (Tue, 22 Jan 2013)
Log Message:
-----------
Enabled config option to apply QTL on a local file with relativ path.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java
Modified: trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java 2013-01-22 12:44:46 UTC (rev 3886)
+++ trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java 2013-01-22 13:21:59 UTC (rev 3887)
@@ -1,5 +1,6 @@
package org.dllearner.kb;
+import java.io.File;
import java.io.InputStream;
import java.net.MalformedURLException;
import java.net.URL;
@@ -18,6 +19,7 @@
private OntModel model;
private String fileName;
+ private String baseDir;
public LocalModelBasedSparqlEndpointKS() {
}
@@ -34,6 +36,14 @@
return fileName;
}
+ public String getBaseDir() {
+ return baseDir;
+ }
+
+ public void setBaseDir(String baseDir) {
+ this.baseDir = baseDir;
+ }
+
public LocalModelBasedSparqlEndpointKS(URL ontologyURL) {
this.fileName = ontologyURL.toString();
}
@@ -44,9 +54,9 @@
@Override
public void init() throws ComponentInitException {
- Model baseModel = ModelFactory.createDefaultModel();
+ Model baseModel = ModelFactory.createDefaultModel();System.out.println("Base: " + baseDir);
// use the FileManager to find the input file
- InputStream in = FileManager.get().open(fileName);
+ InputStream in = FileManager.get().open(baseDir + File.separator + fileName);
if (in == null) {
throw new IllegalArgumentException(
"File: " + fileName + " not found");
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
|
From: <lor...@us...> - 2013-01-22 12:44:54
|
Revision: 3886
http://dl-learner.svn.sourceforge.net/dl-learner/?rev=3886&view=rev
Author: lorenz_b
Date: 2013-01-22 12:44:46 +0000 (Tue, 22 Jan 2013)
Log Message:
-----------
Enabled config option to apply QTL on a local file.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java
trunk/components-core/src/main/java/org/dllearner/core/AnnComponentManager.java
trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java 2013-01-22 11:27:57 UTC (rev 3885)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/qtl/QTL.java 2013-01-22 12:44:46 UTC (rev 3886)
@@ -53,6 +53,7 @@
import org.dllearner.core.options.ConfigOption;
import org.dllearner.core.options.IntegerConfigOption;
import org.dllearner.core.owl.Individual;
+import org.dllearner.kb.LocalModelBasedSparqlEndpointKS;
import org.dllearner.kb.SparqlEndpointKS;
import org.dllearner.kb.sparql.CachingConciseBoundedDescriptionGenerator;
import org.dllearner.kb.sparql.ConciseBoundedDescriptionGenerator;
@@ -69,6 +70,7 @@
import com.hp.hpl.jena.query.ResultSetRewindable;
import com.hp.hpl.jena.rdf.model.Model;
import com.hp.hpl.jena.rdf.model.Statement;
+import com.hp.hpl.jena.rdf.model.StmtIterator;
import com.hp.hpl.jena.util.iterator.Filter;
/**
@@ -112,6 +114,8 @@
private QueryTree<String> lgg;
private SortedSet<String> lggInstances;
+
+ private Set<String> objectNamespacesToIgnore = new HashSet<String>();
public static Collection<ConfigOption<?>> createConfigOptions() {
Collection<ConfigOption<?>> options = new LinkedList<ConfigOption<?>>();
@@ -208,9 +212,13 @@
public void setMaxQueryTreeDepth(int maxQueryTreeDepth){
this.maxQueryTreeDepth = maxQueryTreeDepth;
- cbdGenerator.setRecursionDepth(maxQueryTreeDepth);
+// cbdGenerator.setRecursionDepth(maxQueryTreeDepth);
}
+ public int getMaxQueryTreeDepth() {
+ return maxQueryTreeDepth;
+ }
+
public String getSPARQLQuery(){
if(lgg == null){
lgg = lggGenerator.getLGG(getQueryTrees(posExamples));
@@ -218,6 +226,10 @@
return lgg.toSPARQLQueryString();
}
+ public void setObjectNamespacesToIgnore(Set<String> namespacesToIgnore){
+ this.objectNamespacesToIgnore = namespacesToIgnore;
+ }
+
public void setRestrictToNamespaces(List<String> namespaces){
cbdGenerator.setRestrictToNamespaces(namespaces);
}
@@ -237,12 +249,12 @@
Model model;
QueryTree<String> tree;
for(String resource : resources){
- if(logger.isDebugEnabled()){
- logger.debug("Tree for resource " + resource);
- }
+ logger.info("Generating tree for " + resource);
model = cbdGenerator.getConciseBoundedDescription(resource);
+ applyFilters(model);
tree = treeCache.getQueryTree(resource, model);
if(logger.isDebugEnabled()){
+ logger.debug("Tree for resource " + resource);
logger.debug(tree.getStringRepresentation());
}
trees.add(tree);
@@ -250,6 +262,19 @@
return trees;
}
+ private void applyFilters(Model model){
+ Statement st;
+ for(StmtIterator iter = model.listStatements(); iter.hasNext();){
+ st = iter.next();
+ for(String ns : objectNamespacesToIgnore){
+ if(st.getObject().isURIResource() && st.getObject().asResource().getURI().startsWith(ns)){
+ iter.remove();
+ break;
+ }
+ }
+ }
+ }
+
private List<String> getKnownResources(){
return ListUtils.union(posExamples, negExamples);
}
@@ -305,7 +330,7 @@
if(logger.isDebugEnabled()){
logger.debug("LGG: \n" + lgg.getStringRepresentation());
}
-
+ logger.info(lgg.toSPARQLQuery());
}
@Override
@@ -325,10 +350,14 @@
this.posExamples = convert(((PosNegLP)learningProblem).getPositiveExamples());
this.negExamples = convert(((PosNegLP)learningProblem).getNegativeExamples());
}
- endpoint = endpointKS.getEndpoint();
+ treeCache = new QueryTreeCache();
- treeCache = new QueryTreeCache();
- cbdGenerator = new CachingConciseBoundedDescriptionGenerator(new ConciseBoundedDescriptionGeneratorImpl(endpoint, cache));
+ if(endpointKS instanceof LocalModelBasedSparqlEndpointKS){
+ cbdGenerator = new CachingConciseBoundedDescriptionGenerator(new ConciseBoundedDescriptionGeneratorImpl(((LocalModelBasedSparqlEndpointKS) endpointKS).getModel()));
+ } else {
+ endpoint = endpointKS.getEndpoint();
+ cbdGenerator = new CachingConciseBoundedDescriptionGenerator(new ConciseBoundedDescriptionGeneratorImpl(endpoint, cache));
+ }
cbdGenerator.setRecursionDepth(maxQueryTreeDepth);
lggGenerator = new LGGGeneratorImpl<String>();
Modified: trunk/components-core/src/main/java/org/dllearner/core/AnnComponentManager.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/core/AnnComponentManager.java 2013-01-22 11:27:57 UTC (rev 3885)
+++ trunk/components-core/src/main/java/org/dllearner/core/AnnComponentManager.java 2013-01-22 12:44:46 UTC (rev 3886)
@@ -74,10 +74,11 @@
"org.dllearner.algorithms.properties.SubDataPropertyOfAxiomLearner",
"org.dllearner.algorithms.DisjointClassesLearner",
"org.dllearner.algorithms.SimpleSubclassLearner",
-// "org.dllearner.algorithm.qtl.QTL",
+ "org.dllearner.algorithms.qtl.QTL",
"org.dllearner.kb.KBFile",
"org.dllearner.kb.OWLFile",
"org.dllearner.kb.SparqlEndpointKS",
+ "org.dllearner.kb.LocalModelBasedSparqlEndpointKS",
"org.dllearner.kb.sparql.SparqlKnowledgeSource",
"org.dllearner.kb.sparql.simple.SparqlSimpleExtractor",
"org.dllearner.learningproblems.PosNegLPStandard",
Modified: trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java 2013-01-22 11:27:57 UTC (rev 3885)
+++ trunk/components-core/src/main/java/org/dllearner/kb/LocalModelBasedSparqlEndpointKS.java 2013-01-22 12:44:46 UTC (rev 3886)
@@ -4,6 +4,7 @@
import java.net.MalformedURLException;
import java.net.URL;
+import org.dllearner.core.ComponentAnn;
import org.dllearner.core.ComponentInitException;
import com.hp.hpl.jena.ontology.OntModel;
@@ -12,25 +13,43 @@
import com.hp.hpl.jena.rdf.model.ModelFactory;
import com.hp.hpl.jena.util.FileManager;
+@ComponentAnn(name = "Local Endpoint", shortName = "local_sparql", version = 0.9)
public class LocalModelBasedSparqlEndpointKS extends SparqlEndpointKS {
private OntModel model;
+ private String fileName;
- public LocalModelBasedSparqlEndpointKS(OntModel model) {
- this.model = model;
+ public LocalModelBasedSparqlEndpointKS() {
}
public LocalModelBasedSparqlEndpointKS(String ontologyURL) throws MalformedURLException {
this(new URL(ontologyURL));
}
+ public void setFileName(String fileName) {
+ this.fileName = fileName;
+ }
+
+ public String getFileName() {
+ return fileName;
+ }
+
public LocalModelBasedSparqlEndpointKS(URL ontologyURL) {
+ this.fileName = ontologyURL.toString();
+ }
+
+ public LocalModelBasedSparqlEndpointKS(OntModel model) {
+ this.model = model;
+ }
+
+ @Override
+ public void init() throws ComponentInitException {
Model baseModel = ModelFactory.createDefaultModel();
// use the FileManager to find the input file
- InputStream in = FileManager.get().open(ontologyURL.toString());
+ InputStream in = FileManager.get().open(fileName);
if (in == null) {
throw new IllegalArgumentException(
- "File: " + ontologyURL + " not found");
+ "File: " + fileName + " not found");
}
// read the RDF/XML file
baseModel.read(in, null);
@@ -38,11 +57,6 @@
model = ModelFactory.createOntologyModel(OntModelSpec.RDFS_MEM_RDFS_INF, baseModel);
}
- @Override
- public void init() throws ComponentInitException {
-
- }
-
public OntModel getModel() {
return model;
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
3 messages has been excluded from this view by a project administrator.
