Menu
▾
▴
[DL-Learner SVN] SF.net SVN: dl-learner:[4199] trunk/components-core/src/main/java/org/ dllearner/algorithms/isle/index
|
From: <dfl...@us...> - 2013-12-09 15:37:42
|
Revision: 4199
http://sourceforge.net/p/dl-learner/code/4199
Author: dfleischhacker
Date: 2013-12-09 15:37:39 +0000 (Mon, 09 Dec 2013)
Log Message:
-----------
Cleanup and show alternative names
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/Token.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TrieEntityCandidateGenerator.java
Removed Paths:
-------------
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/FullTokenEntitySetPair.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/NormalizedTextMapper.java
Deleted: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/FullTokenEntitySetPair.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/FullTokenEntitySetPair.java 2013-12-09 15:36:38 UTC (rev 4198)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/FullTokenEntitySetPair.java 2013-12-09 15:37:39 UTC (rev 4199)
@@ -1,31 +0,0 @@
-package org.dllearner.algorithms.isle.index;
-
-import org.dllearner.core.owl.Entity;
-
-import java.util.HashSet;
-import java.util.Set;
-
-/**
- * A pair consisting of a full string token and the corresponding entities
- */
-public class FullTokenEntitySetPair {
- private String fullToken;
- private Set<Entity> entitySet;
-
- public FullTokenEntitySetPair(String fullToken) {
- this.fullToken = fullToken;
- this.entitySet = new HashSet<Entity>();
- }
-
- public String getFullToken() {
- return fullToken;
- }
-
- public Set<Entity> getEntitySet() {
- return entitySet;
- }
-
- public void addEntity(Entity entity) {
- entitySet.add(entity);
- }
-}
Deleted: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/NormalizedTextMapper.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/NormalizedTextMapper.java 2013-12-09 15:36:38 UTC (rev 4198)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/NormalizedTextMapper.java 2013-12-09 15:37:39 UTC (rev 4199)
@@ -1,141 +0,0 @@
-package org.dllearner.algorithms.isle.index;
-
-import java.util.ArrayList;
-
-/**
- * Provides text normalization and mapping of normalized ranges to the original ones.
- */
-public class NormalizedTextMapper {
- private Document originalDocument;
- private String originalText;
- private String normalizedText;
-
- private ArrayList<OccurenceMappingPair> normalizedIndexToOriginalIndex;
-
- public NormalizedTextMapper(Document original) {
- this.originalDocument = original;
- this.originalText = original.getContent();
- this.normalizedIndexToOriginalIndex = new ArrayList<OccurenceMappingPair>();
-
- StringBuilder sb = new StringBuilder();
- int currentOriginalIndex = 0;
- for (String originalWord : originalText.split(" ")) {
- String normalizedWord = getNormalizedWord(originalWord);
- normalizedIndexToOriginalIndex
- .add(new OccurenceMappingPair(currentOriginalIndex, originalWord.length(), sb.length(),
- normalizedWord.length()));
- currentOriginalIndex += originalWord.length() + 1;
- sb.append(normalizedWord);
- sb.append(" ");
- }
- normalizedText = sb.toString();
- }
-
- public String getOriginalText() {
- return originalText;
- }
-
- public String getNormalizedText() {
- return normalizedText;
- }
-
- /**
- * Returns the annotation for the original text matching the given position and length in the normalized
- * text.
- *
- * @param position position in the normalized text to get annotation for
- * @param length length of the text to get annotation for
- * @return
- */
- public Annotation getOriginalAnnotationForPosition(int position, int length) {
- int curNormalizedLength = 0;
- int originalStart = -1;
- int curOriginalLength = 0;
-
- for (OccurenceMappingPair p : normalizedIndexToOriginalIndex) {
- if (p.getNormalizedIndex() == position) {
- originalStart = p.getOriginalIndex();
- }
- if (originalStart != -1) {
- curNormalizedLength += p.getNormalizedLength();
- curOriginalLength += p.getOriginalLength();
- if (curNormalizedLength >= length) {
- //TODO refactoring
-// return new Annotation(originalDocument, originalStart, curOriginalLength);
- }
-
- // include space
- curNormalizedLength += 1;
- curOriginalLength += 1;
- }
- }
-
- return null;
- }
-
- /**
- * Returns the normalized form of the given word. Word must not contain any spaces or the like.
- * @param word
- * @return
- */
- private String getNormalizedWord(String word) {
- return LinguisticUtil.getInstance().getNormalizedForm(word);
- }
-
- public static void main(String[] args) {
-// NormalizedTextMapper n = new NormalizedTextMapper(new TextDocument("This is a testing text using letters"));
-// System.out.println(n.getOriginalText());
-// System.out.println(n.getNormalizedText());
-// for (OccurenceMappingPair p : n.normalizedIndexToOriginalIndex) {
-// System.out.println(p);
-// }
-// System.out.println(n.getOriginalAnnotationForPosition(7,6));
-// System.out.println(n.getOriginalAnnotationForPosition(23,6));
-// System.out.println(n.getOriginalAnnotationForPosition(7,1));
-// System.out.println(n.getOriginalAnnotationForPosition(14,15));
- }
-
- /**
- * Maps words identified by index and length in the normalized texts to the original word.
- */
- private class OccurenceMappingPair {
- private int originalIndex;
- private int originalLength;
- private int normalizedIndex;
- private int normalizedLength;
-
- private OccurenceMappingPair(int originalIndex, int originalLength, int normalizedIndex, int normalizedLength) {
-
- this.originalIndex = originalIndex;
- this.originalLength = originalLength;
- this.normalizedIndex = normalizedIndex;
- this.normalizedLength = normalizedLength;
- }
-
- private int getNormalizedIndex() {
- return normalizedIndex;
- }
-
- private int getNormalizedLength() {
- return normalizedLength;
- }
-
- private int getOriginalLength() {
- return originalLength;
- }
-
- private int getOriginalIndex() {
- return originalIndex;
- }
-
- @Override
- public String toString() {
- return "OccurenceMappingPair{" +
- "originalIndex=" + originalIndex +
- ", originalLength=" + originalLength +
- ", normalizedIndex=" + normalizedIndex +
- ", normalizedLength=" + normalizedLength +
- '}';
- }
- }
-}
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 2013-12-09 15:36:38 UTC (rev 4198)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/SimpleEntityCandidatesTrie.java 2013-12-09 15:37:39 UTC (rev 4199)
@@ -91,6 +91,10 @@
String[] synonyms = LinguisticUtil.getInstance().getSynonymsForWord(t.getRawForm(), wordnetPos);
for (String synonym : synonyms) {
+ // ignore all multi word synonyms
+ if (synonym.contains("_")) {
+ continue;
+ }
t.addAlternativeForm(LinguisticUtil.getInstance().getNormalizedForm(synonym));
}
}
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/Token.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/Token.java 2013-12-09 15:36:38 UTC (rev 4198)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/Token.java 2013-12-09 15:37:39 UTC (rev 4199)
@@ -138,7 +138,7 @@
*/
@Override
public String toString() {
- return "[Word: " + rawForm + " | Stemmed word: " + stemmedForm + " | POS tag: " + posTag + "]";
+ return "[Word: " + rawForm + " | Stemmed word: " + stemmedForm + " | POS tag: " + posTag + " | Alternatives: " + alternativeForms.toString() + "]";
}
/**
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 2013-12-09 15:36:38 UTC (rev 4198)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/TrieEntityCandidateGenerator.java 2013-12-09 15:37:39 UTC (rev 4199)
@@ -1,15 +1,15 @@
package org.dllearner.algorithms.isle.index;
import com.google.common.collect.Lists;
-import com.google.common.collect.Sets;
-
import org.dllearner.algorithms.isle.EntityCandidateGenerator;
import org.dllearner.algorithms.isle.StopWordFilter;
import org.dllearner.core.owl.Entity;
import org.semanticweb.owlapi.model.OWLOntology;
-import java.util.*;
-import java.util.regex.Pattern;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Set;
/**
* Generates candidates using a entity candidates prefix trie
@@ -34,7 +34,6 @@
/**
* Postprocess the annotations generated by annotate
* The objective is to merge annotations which are likely to belong to the same entity
- * @param annotations : set of annotations
* @param window : maximum distance between the annotations
* @return
*/
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
