Menu
▾
▴
[DL-Learner SVN] SF.net SVN: dl-learner:[4183] trunk/components-core/src/main/java/org/ dllearner/algorithms/isle
|
From: <lor...@us...> - 2013-12-02 14:52:35
|
Revision: 4183
http://sourceforge.net/p/dl-learner/code/4183
Author: lorenz_b
Date: 2013-12-02 14:52:33 +0000 (Mon, 02 Dec 2013)
Log Message:
-----------
Refactoring.
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/SimpleEntityCandidatesTrie.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TokenTree.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/AnnotationEntityTextRetriever.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/EntityTextRetriever.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/RDFSLabelEntityTextRetriever.java
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/SimpleEntityCandidatesTrie.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/SimpleEntityCandidatesTrie.java 2013-12-02 14:41:21 UTC (rev 4182)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/SimpleEntityCandidatesTrie.java 2013-12-02 14:52:33 UTC (rev 4183)
@@ -42,26 +42,21 @@
public void buildTrie(OWLOntology ontology, NameGenerator nameGenerator) {
this.trie = new PrefixTrie<FullTokenEntitySetPair>();
- Map<Entity, Set<String>> relevantText = entityTextRetriever.getRelevantText(ontology);
+ Map<Entity, Set<List<Token>>> entity2TokenSet = entityTextRetriever.getRelevantText(ontology);
- for (Entity entity : relevantText.keySet()) {
-
- for (String text : relevantText.get(entity)) {
- text = StringUtils.join(LinguisticUtil.getInstance().getWordsFromCamelCase(text), " ");
- text = StringUtils.join(LinguisticUtil.getInstance().getWordsFromUnderscored(text), " ");
- if (text.trim().isEmpty()) {
- continue;
- }
-
- addEntry(text, entity);
- addSubsequencesWordNet(entity, text);
-
- for (String alternativeText : nameGenerator.getAlternativeText(text)) {
- addEntry(alternativeText.toLowerCase(), entity, text);
- }
- }
- }
+ for (Entry<Entity, Set<List<Token>>> entry : entity2TokenSet.entrySet()) {
+ Entity entity = entry.getKey();
+ Set<List<Token>> tokenSet = entry.getValue();
+ for (List<Token> tokens : tokenSet) {
+ addEntry(tokens, entity);
+ addSubsequences(entity, tokens);
+// addSubsequencesWordNet(entity, text);
+// for (String alternativeText : nameGenerator.getAlternativeText(text)) {
+// addEntry(alternativeText.toLowerCase(), entity, text);
+// }
+ }
+ }
}
/**
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TokenTree.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TokenTree.java 2013-12-02 14:41:21 UTC (rev 4182)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TokenTree.java 2013-12-02 14:52:33 UTC (rev 4183)
@@ -127,6 +127,15 @@
tree.add(tokens1, new NamedClass("TokenTree"));
tree.add(tokens2, new NamedClass("TokenizedTree"));
System.out.println(tree);
+
+ System.out.println(tree.getEntitiesForLongestMatch(tokens1));
+ System.out.println(tree.getLongestMatch(tokens1));
+
+ List<Token> tokens3 = Lists.newLinkedList();
+ for (String s : Splitter.on(" ").split("this is a very nice tokenized tree")) {
+ tokens3.add(new Token(s, s, s, false, false));
+ };
+ System.out.println(tree.getLongestMatch(tokens3));
}
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/AnnotationEntityTextRetriever.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/AnnotationEntityTextRetriever.java 2013-12-02 14:41:21 UTC (rev 4182)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/AnnotationEntityTextRetriever.java 2013-12-02 14:52:33 UTC (rev 4183)
@@ -5,10 +5,13 @@
import java.util.HashMap;
import java.util.HashSet;
+import java.util.List;
import java.util.Map;
import java.util.Set;
+import org.dllearner.algorithms.isle.TextDocumentGenerator;
import org.dllearner.algorithms.isle.index.LinguisticUtil;
+import org.dllearner.algorithms.isle.index.Token;
import org.dllearner.core.owl.Entity;
import org.dllearner.kb.OWLAPIOntology;
import org.dllearner.utilities.owl.OWLAPIConverter;
@@ -75,8 +78,8 @@
* @see org.dllearner.algorithms.isle.EntityTextRetriever#getRelevantText(org.dllearner.core.owl.Entity)
*/
@Override
- public Map<String, Double> getRelevantText(Entity entity) {
- Map<String, Double> textWithWeight = new HashMap<String, Double>();
+ public Map<List<Token>, Double> getRelevantText(Entity entity) {
+ Map<List<Token>, Double> textWithWeight = new HashMap<List<Token>, Double>();
OWLEntity e = OWLAPIConverter.getOWLAPIEntity(entity);
@@ -87,7 +90,7 @@
OWLLiteral val = (OWLLiteral) annotation.getValue();
if (val.hasLang(language)) {
String label = val.getLiteral().trim();
- textWithWeight.put(label, weight);
+ textWithWeight.put(TextDocumentGenerator.getInstance().generateDocument(label), weight);
}
}
}
@@ -97,7 +100,7 @@
String shortForm = sfp.getShortForm(IRI.create(entity.getURI()));
shortForm = Joiner.on(" ").join(LinguisticUtil.getInstance().getWordsFromCamelCase(shortForm));
shortForm = Joiner.on(" ").join(LinguisticUtil.getInstance().getWordsFromUnderscored(shortForm)).trim();
- textWithWeight.put(shortForm, weight);
+ textWithWeight.put(TextDocumentGenerator.getInstance().generateDocument(shortForm), weight);
}
return textWithWeight;
@@ -108,8 +111,8 @@
* @return
*/
@Override
- public Map<Entity, Set<String>> getRelevantText(OWLOntology ontology) {
- Map<Entity, Set<String>> entity2RelevantText = new HashMap<Entity, Set<String>>();
+ public Map<Entity, Set<List<Token>>> getRelevantText(OWLOntology ontology) {
+ Map<Entity, Set<List<Token>>> entity2RelevantText = new HashMap<>();
Set<OWLEntity> schemaEntities = new HashSet<OWLEntity>();
schemaEntities.addAll(ontology.getClassesInSignature());
@@ -117,7 +120,7 @@
schemaEntities.addAll(ontology.getDataPropertiesInSignature());
schemaEntities.remove(OWL_THING);
- Map<String, Double> relevantText;
+ Map<List<Token>, Double> relevantText;
for (OWLEntity owlEntity : schemaEntities) {
Entity entity = OWLAPIConverter.getEntity(owlEntity);
relevantText = getRelevantText(entity);
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/EntityTextRetriever.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/EntityTextRetriever.java 2013-12-02 14:41:21 UTC (rev 4182)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/EntityTextRetriever.java 2013-12-02 14:52:33 UTC (rev 4183)
@@ -19,9 +19,11 @@
package org.dllearner.algorithms.isle.textretrieval;
+import java.util.List;
import java.util.Map;
import java.util.Set;
+import org.dllearner.algorithms.isle.index.Token;
import org.dllearner.core.owl.Entity;
import org.semanticweb.owlapi.model.OWLOntology;
@@ -45,8 +47,8 @@
* @param entity The entity to handle.
* @return A weighted set of strings. For a value x, we need to have 0 <= x <= 1.
*/
- public Map<String, Double> getRelevantText(Entity entity);
+ public Map<List<Token>, Double> getRelevantText(Entity entity);
- public Map<Entity, Set<String>> getRelevantText(OWLOntology ontology);
+ public Map<Entity, Set<List<Token>>> getRelevantText(OWLOntology ontology);
}
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/RDFSLabelEntityTextRetriever.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/RDFSLabelEntityTextRetriever.java 2013-12-02 14:41:21 UTC (rev 4182)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/textretrieval/RDFSLabelEntityTextRetriever.java 2013-12-02 14:52:33 UTC (rev 4183)
@@ -4,12 +4,14 @@
package org.dllearner.algorithms.isle.textretrieval;
import java.io.File;
+import java.util.List;
import java.util.Map;
import java.util.Map.Entry;
import java.util.Set;
import java.util.SortedMap;
import java.util.TreeMap;
+import org.dllearner.algorithms.isle.index.Token;
import org.dllearner.core.owl.Entity;
import org.dllearner.kb.OWLAPIOntology;
import org.semanticweb.owlapi.apibinding.OWLManager;
@@ -43,13 +45,13 @@
OWLOntology ontology = man.loadOntology(IRI.create("http://www.semanticbible.com/2006/11/NTNames.owl"));
RDFSLabelEntityTextRetriever labelRetriever = new RDFSLabelEntityTextRetriever(ontology);
- Map<Entity, Set<String>> relevantText = labelRetriever.getRelevantText(ontology);
+ Map<Entity, Set<List<Token>>> relevantText = labelRetriever.getRelevantText(ontology);
SortedMap<String, String> uri2Labels = new TreeMap<String, String>();
- for (Entry<Entity, Set<String>> entry : relevantText.entrySet()) {
+ for (Entry<Entity, Set<List<Token>>> entry : relevantText.entrySet()) {
Entity key = entry.getKey();
- Set<String> value = entry.getValue();
- uri2Labels.put(key.getName(), value.iterator().next());
+ Set<List<Token>> value = entry.getValue();
+ uri2Labels.put(key.getName(), value.iterator().next().get(0).getRawForm());
}
StringBuilder csv = new StringBuilder();
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
