Menu
▾
▴
[DL-Learner SVN] SF.net SVN: dl-learner:[4163] trunk/components-core/src/main/java/org/ dllearner/algorithms/isle
|
From: <dfl...@us...> - 2013-11-21 13:16:16
|
Revision: 4163
http://sourceforge.net/p/dl-learner/code/4163
Author: dfleischhacker
Date: 2013-11-21 13:16:13 +0000 (Thu, 21 Nov 2013)
Log Message:
-----------
Annotation refactoring
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/Annotation.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TrieEntityCandidateGenerator.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/semantic/SemanticIndex.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/wsd/SentenceBasedContextExtractor.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/wsd/SimpleWordSenseDisambiguation.java
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/Annotation.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/Annotation.java 2013-11-21 13:00:33 UTC (rev 4162)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/Annotation.java 2013-11-21 13:16:13 UTC (rev 4163)
@@ -4,6 +4,9 @@
package org.dllearner.algorithms.isle.index;
+import java.util.ArrayList;
+import java.util.List;
+
/**
* A (non-semantic) annotation which represents an entity in a document by its offset and length.
* @author Lorenz Buehmann
@@ -12,8 +15,7 @@
public class Annotation {
private Document referencedDocument;
- private int offset;
- private int length;
+ private ArrayList<Token> tokens;
private String matchedString;
public String getMatchedString() {
@@ -24,64 +26,64 @@
this.matchedString = matchedString;
}
- public Annotation(Document referencedDocument, int offset, int length) {
+ public Annotation(Document referencedDocument, List<Token> tokens) {
this.referencedDocument = referencedDocument;
- this.offset = offset;
- this.length = length;
- }
+ this.tokens = new ArrayList<Token>(tokens);
+ }
public Document getReferencedDocument() {
return referencedDocument;
}
- public int getOffset() {
- return offset;
- }
+ public String getString(){
+ StringBuilder sb = new StringBuilder();
+ for (Token t : tokens) {
+ if (sb.length() > 0) {
+ sb.append(" ");
+ }
+ sb.append(t.getStemmedForm());
+ }
+ return sb.toString();
+ }
- public int getLength() {
- return length;
- }
-
- public String getToken(){
- return referencedDocument.getContent().substring(offset, offset + length);
- }
+ @Override
+ public boolean equals(Object o) {
+ if (this == o) {
+ return true;
+ }
+ if (o == null || getClass() != o.getClass()) {
+ return false;
+ }
- @Override
- public int hashCode() {
- final int prime = 31;
- int result = 1;
- result = prime * result + ((referencedDocument == null) ? 0 : referencedDocument.hashCode());
- result = prime * result + length;
- result = prime * result + offset;
- return result;
- }
+ Annotation that = (Annotation) o;
+ if (matchedString != null ? !matchedString.equals(that.matchedString) : that.matchedString != null) {
+ return false;
+ }
+ if (referencedDocument != null ? !referencedDocument.equals(that.referencedDocument) :
+ that.referencedDocument != null) {
+ return false;
+ }
+ if (tokens != null ? !tokens.equals(that.tokens) : that.tokens != null) {
+ return false;
+ }
+
+ return true;
+ }
+
+ @Override
+ public int hashCode() {
+ int result = referencedDocument != null ? referencedDocument.hashCode() : 0;
+ result = 31 * result + (tokens != null ? tokens.hashCode() : 0);
+ result = 31 * result + (matchedString != null ? matchedString.hashCode() : 0);
+ return result;
+ }
+
+ /* (non-Javadoc)
+ * @see java.lang.Object#toString()
+ */
@Override
- public boolean equals(Object obj) {
- if (this == obj)
- return true;
- if (obj == null)
- return false;
- if (getClass() != obj.getClass())
- return false;
- Annotation other = (Annotation) obj;
- if (referencedDocument == null) {
- if (other.referencedDocument != null)
- return false;
- } else if (!referencedDocument.equals(other.referencedDocument))
- return false;
- if (length != other.length)
- return false;
- if (offset != other.offset)
- return false;
- return true;
- }
-
- /* (non-Javadoc)
- * @see java.lang.Object#toString()
- */
- @Override
public String toString() {
- return "\"" + referencedDocument.getContent().substring(offset, offset+length) + "\" at position " + offset;
- }
+ return getString();
+ }
}
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TrieEntityCandidateGenerator.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TrieEntityCandidateGenerator.java 2013-11-21 13:00:33 UTC (rev 4162)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TrieEntityCandidateGenerator.java 2013-11-21 13:16:13 UTC (rev 4163)
@@ -54,7 +54,7 @@
Annotation annotation_i = sortedAnnotations.get(i);
int begin_i = annotation_i.getOffset();
int end_i = begin_i + annotation_i.getLength()-1;
- String token_i = annotation_i.getToken();
+ String token_i = annotation_i.getString();
Set<Entity> candidates_i = getCandidates(annotation_i);
Set<Entity> newCandidates_i = new HashSet<Entity>();
@@ -68,7 +68,7 @@
for (int j=windowStart; j<sortedAnnotations.size() && j<windowEnd; j++) {
if (j!=i) {
Annotation annotation_j = sortedAnnotations.get(j);
- String token_j = annotation_j.getToken();
+ String token_j = annotation_j.getString();
Set<Entity> candidates_j = getCandidates(annotation_j);
Set<Entity> intersection = Sets.intersection(candidates_i, candidates_j);
Set<Entity> newCandidates_ij = new HashSet<Entity>();
@@ -83,7 +83,7 @@
if (!newCandidates_ij.isEmpty()) {
Annotation mergedAnnotation = mergeAnnotations(annotation_i,annotation_j);
// If there's no punctuation in the merged annotation
- if (!Pattern.matches("\\p{Punct}", mergedAnnotation.getToken())) {
+ if (!Pattern.matches("\\p{Punct}", mergedAnnotation.getString())) {
candidatesMap.put(mergedAnnotation, newCandidates_ij);
candidatesMap.remove(annotation_i);
candidatesMap.remove(annotation_j);
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/semantic/SemanticIndex.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/semantic/SemanticIndex.java 2013-11-21 13:00:33 UTC (rev 4162)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/semantic/SemanticIndex.java 2013-11-21 13:16:13 UTC (rev 4163)
@@ -48,6 +48,7 @@
public SemanticIndex(OWLOntology ontology) {
this.ontology = ontology;
+
}
/**
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/wsd/SentenceBasedContextExtractor.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/wsd/SentenceBasedContextExtractor.java 2013-11-21 13:00:33 UTC (rev 4162)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/wsd/SentenceBasedContextExtractor.java 2013-11-21 13:16:13 UTC (rev 4163)
@@ -57,7 +57,7 @@
}
index += s.length();
}
- throw new RuntimeException("Token " + annotation.getToken() + " not found in text " + annotation.getReferencedDocument().getRawContent());
+ throw new RuntimeException("Token " + annotation.getString() + " not found in text " + annotation.getReferencedDocument().getRawContent());
}
private List<CoreMap> getSentences(String document) {
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/wsd/SimpleWordSenseDisambiguation.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/wsd/SimpleWordSenseDisambiguation.java 2013-11-21 13:00:33 UTC (rev 4162)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/wsd/SimpleWordSenseDisambiguation.java 2013-11-21 13:16:13 UTC (rev 4163)
@@ -50,7 +50,7 @@
public SemanticAnnotation disambiguate(Annotation annotation, Set<Entity> candidateEntities) {
logger.debug("Linguistic annotations:\n" + annotation);
logger.debug("Candidate entities:" + candidateEntities);
- String token = annotation.getToken().trim();
+ String token = annotation.getString().trim();
//check if annotation token matches label of entity or the part behind #(resp. /)
for (Entity entity : candidateEntities) {
Set<String> labels = getLabels(entity);
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
