Menu
▾
▴
[DL-Learner SVN] SF.net SVN: dl-learner:[4092] trunk/components-core/src/main/java/org/ dllearner/algorithms/isle
|
From: <dfl...@us...> - 2013-09-06 10:01:56
|
Revision: 4092
http://sourceforge.net/p/dl-learner/code/4092
Author: dfleischhacker
Date: 2013-09-06 10:01:53 +0000 (Fri, 06 Sep 2013)
Log Message:
-----------
Add methods to get top n synonyms for words
Modified Paths:
--------------
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/WordNet.java
trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/LinguisticUtil.java
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/WordNet.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/WordNet.java 2013-09-05 13:59:47 UTC (rev 4091)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/WordNet.java 2013-09-06 10:01:53 UTC (rev 4092)
@@ -1,78 +1,115 @@
package org.dllearner.algorithms.isle;
-import java.io.InputStream;
-import java.util.ArrayList;
-import java.util.Iterator;
-import java.util.List;
-import java.util.Set;
-
import net.didion.jwnl.JWNL;
import net.didion.jwnl.JWNLException;
-import net.didion.jwnl.data.IndexWord;
-import net.didion.jwnl.data.POS;
-import net.didion.jwnl.data.PointerTarget;
-import net.didion.jwnl.data.PointerUtils;
-import net.didion.jwnl.data.Synset;
-import net.didion.jwnl.data.Word;
+import net.didion.jwnl.data.*;
import net.didion.jwnl.data.list.PointerTargetNode;
import net.didion.jwnl.data.list.PointerTargetNodeList;
import net.didion.jwnl.dictionary.Dictionary;
+import java.io.InputStream;
+import java.util.ArrayList;
+import java.util.Iterator;
+import java.util.List;
+
public class WordNet {
-
- public Dictionary dict;
-
- public WordNet() {
- try {
- JWNL.initialize(this.getClass().getClassLoader().getResourceAsStream("wordnet_properties.xml"));
- dict = Dictionary.getInstance();
- } catch (JWNLException e) {
- e.printStackTrace();
- }
- }
-
- public WordNet(String configPath) {
- try {
- JWNL.initialize(this.getClass().getClassLoader().getResourceAsStream(configPath));
- dict = Dictionary.getInstance();
- } catch (JWNLException e) {
- e.printStackTrace();
- }
- }
-
- public WordNet(InputStream propertiesStream) {
- try {
- JWNL.initialize(propertiesStream);
- dict = Dictionary.getInstance();
- } catch (JWNLException e) {
- e.printStackTrace();
- }
- }
-
- public List<String> getBestSynonyms(POS pos, String s) {
-
- List<String> synonyms = new ArrayList<String>();
-
- try {
- IndexWord iw = dict.getIndexWord(pos, s);//dict.getMorphologicalProcessor().lookupBaseForm(pos, s)
+
+ public Dictionary dict;
+
+ public WordNet() {
+ try {
+ JWNL.initialize(this.getClass().getClassLoader().getResourceAsStream("wordnet_properties.xml"));
+ dict = Dictionary.getInstance();
+ }
+ catch (JWNLException e) {
+ e.printStackTrace();
+ }
+ }
+
+ public WordNet(String configPath) {
+ try {
+ JWNL.initialize(this.getClass().getClassLoader().getResourceAsStream(configPath));
+ dict = Dictionary.getInstance();
+ }
+ catch (JWNLException e) {
+ e.printStackTrace();
+ }
+ }
+
+ public WordNet(InputStream propertiesStream) {
+ try {
+ JWNL.initialize(propertiesStream);
+ dict = Dictionary.getInstance();
+ }
+ catch (JWNLException e) {
+ e.printStackTrace();
+ }
+ }
+
+ public static void main(String[] args) {
+ System.out.println(new WordNet().getBestSynonyms(POS.VERB, "learn"));
+ System.out.println(new WordNet().getSisterTerms(POS.NOUN, "actress"));
+ }
+
+ public List<String> getBestSynonyms(POS pos, String s) {
+
+ List<String> synonyms = new ArrayList<String>();
+
+ try {
+ IndexWord iw = dict.getIndexWord(pos, s);//dict.getMorphologicalProcessor().lookupBaseForm(pos, s)
// IndexWord iw = dict.getMorphologicalProcessor().lookupBaseForm(pos, s);
- if(iw != null){
- Synset[] synsets = iw.getSenses();
- Word[] words = synsets[0].getWords();
- for(Word w : words){
- String c = w.getLemma();
- if (!c.equals(s) && !c.contains(" ") && synonyms.size() < 4) {
- synonyms.add(c);
- }
- }
- }
-
- } catch (JWNLException e) {
- e.printStackTrace();
- }
- return synonyms;
- }
+ if (iw != null) {
+ Synset[] synsets = iw.getSenses();
+ Word[] words = synsets[0].getWords();
+ for (Word w : words) {
+ String c = w.getLemma();
+ if (!c.equals(s) && !c.contains(" ") && synonyms.size() < 4) {
+ synonyms.add(c);
+ }
+ }
+ }
+ }
+ catch (JWNLException e) {
+ e.printStackTrace();
+ }
+ return synonyms;
+ }
+
+ /**
+ * Returns the lemmas for the top {@code n} synsets of the given POS for the string {@code s}.
+ *
+ * @param pos the part of speech to retrieve synonyms for
+ * @param s the string to retrieve synonyms for
+ * @param n the number of synonyms to retrieve
+ * @return list of the lemmas of the top n synonyms of s
+ */
+ public List<String> getTopSynonyms(POS pos, String s, int n) {
+
+ List<String> synonyms = new ArrayList<String>();
+
+ try {
+ IndexWord iw = dict.getIndexWord(pos, s);//dict.getMorphologicalProcessor().lookupBaseForm(pos, s)
+// IndexWord iw = dict.getMorphologicalProcessor().lookupBaseForm(pos, s);
+ if (iw != null) {
+ Synset[] synsets = iw.getSenses();
+ for (int i = 0; i < n; i++) {
+ for (Word word : synsets[i].getWords()) {
+ String c = word.getLemma();
+ if (!c.equals(s) && !c.contains(" ")) {
+ synonyms.add(c);
+ }
+ }
+ }
+ }
+
+ }
+ catch (JWNLException e) {
+ e.printStackTrace();
+ }
+ return synonyms;
+ }
+
public List<String> getAllSynonyms(POS pos, String s) {
List<String> synonyms = new ArrayList<String>();
try {
@@ -96,120 +133,124 @@
return synonyms;
}
- public List<String> getSisterTerms(POS pos, String s){
- List<String> sisterTerms = new ArrayList<String>();
-
- try {
- IndexWord iw = dict.getIndexWord(pos, s);//dict.getMorphologicalProcessor().lookupBaseForm(pos, s)
+ public List<String> getSisterTerms(POS pos, String s) {
+ List<String> sisterTerms = new ArrayList<String>();
+
+ try {
+ IndexWord iw = dict.getIndexWord(pos, s);//dict.getMorphologicalProcessor().lookupBaseForm(pos, s)
// IndexWord iw = dict.getMorphologicalProcessor().lookupBaseForm(pos, s);
- if(iw != null){
- Synset[] synsets = iw.getSenses();
- //System.out.println(synsets[0]);
- PointerTarget[] pointerArr = synsets[0].getTargets();
- }
-
- } catch (JWNLException e) {
- e.printStackTrace();
- }
- return sisterTerms;
- }
-
- public List<String> getAttributes(String s) {
-
- List<String> result = new ArrayList<String>();
-
- try {
- IndexWord iw = dict.getIndexWord(POS.ADJECTIVE, s);
- if(iw != null){
- Synset[] synsets = iw.getSenses();
- Word[] words = synsets[0].getWords();
- for(Word w : words){
- String c = w.getLemma();
- if (!c.equals(s) && !c.contains(" ") && result.size() < 4) {
- result.add(c);
- }
- }
- }
-
- } catch (JWNLException e) {
- e.printStackTrace();
- }
-
- return result;
- }
-
- public static void main(String[] args) {
- System.out.println(new WordNet().getBestSynonyms(POS.VERB, "learn"));
- System.out.println(new WordNet().getSisterTerms(POS.NOUN, "actress"));
- }
-
- /**
- * Funktion returns a List of Hypo and Hypernyms of a given string
- * @param s Word for which you want to get Hypo and Hypersyms
- * @return List of Hypo and Hypernyms
- * @throws JWNLException
- */
- public List<String> getRelatedNouns(String s) {
- List<String> result = new ArrayList<String>();
- IndexWord word = null;
- Synset sense=null;
- try{
- word=dict.getIndexWord(POS.NOUN,s);
- if(word!=null){
- sense = word.getSense(1);
- //Synset sense = word.getSense(1);
-
- PointerTargetNodeList relatedListHypernyms = null;
- PointerTargetNodeList relatedListHyponyms = null;
- try {
- relatedListHypernyms = PointerUtils.getInstance().getDirectHypernyms(sense);
- } catch (JWNLException e) {
- // TODO Auto-generated catch block
- e.printStackTrace();
- }
- try {
- relatedListHyponyms = PointerUtils.getInstance().getDirectHyponyms(sense);
- } catch (JWNLException e) {
- // TODO Auto-generated catch block
- e.printStackTrace();
- }
-
- Iterator i = relatedListHypernyms.iterator();
- while (i.hasNext()) {
- PointerTargetNode related = (PointerTargetNode) i.next();
- Synset s1 = related.getSynset();
- String tmp=(s1.toString()).replace(s1.getGloss(), "");
- tmp=tmp.replace(" -- ()]","");
- tmp=tmp.replaceAll("[0-9]","");
- tmp=tmp.replace("[Synset: [Offset: ","");
- tmp=tmp.replace("] [POS: noun] Words: ","");
- //its possible, that there is more than one word in a line from wordnet
- String[] array_tmp=tmp.split(",");
- for(String z : array_tmp) result.add(z.replace(" ", ""));
- }
-
- Iterator j = relatedListHyponyms.iterator();
- while (j.hasNext()) {
- PointerTargetNode related = (PointerTargetNode) j.next();
- Synset s1 = related.getSynset();
- String tmp=(s1.toString()).replace(s1.getGloss(), "");
- tmp=tmp.replace(" -- ()]","");
- tmp=tmp.replaceAll("[0-9]","");
- tmp=tmp.replace("[Synset: [Offset: ","");
- tmp=tmp.replace("] [POS: noun] Words: ","");
- //its possible, that there is more than one word in a line from wordnet
- String[] array_tmp=tmp.split(",");
- for(String z : array_tmp) result.add(z.replace(" ", ""));
- }
- }
- }catch (JWNLException e) {
- // TODO Auto-generated catch block
- e.printStackTrace();
- }
-
-
-
- return result;
- }
-
+ if (iw != null) {
+ Synset[] synsets = iw.getSenses();
+ //System.out.println(synsets[0]);
+ PointerTarget[] pointerArr = synsets[0].getTargets();
+ }
+
+ }
+ catch (JWNLException e) {
+ e.printStackTrace();
+ }
+ return sisterTerms;
+ }
+
+ public List<String> getAttributes(String s) {
+
+ List<String> result = new ArrayList<String>();
+
+ try {
+ IndexWord iw = dict.getIndexWord(POS.ADJECTIVE, s);
+ if (iw != null) {
+ Synset[] synsets = iw.getSenses();
+ Word[] words = synsets[0].getWords();
+ for (Word w : words) {
+ String c = w.getLemma();
+ if (!c.equals(s) && !c.contains(" ") && result.size() < 4) {
+ result.add(c);
+ }
+ }
+ }
+
+ }
+ catch (JWNLException e) {
+ e.printStackTrace();
+ }
+
+ return result;
+ }
+
+ /**
+ * Funktion returns a List of Hypo and Hypernyms of a given string
+ *
+ * @param s Word for which you want to get Hypo and Hypersyms
+ * @return List of Hypo and Hypernyms
+ * @throws JWNLException
+ */
+ public List<String> getRelatedNouns(String s) {
+ List<String> result = new ArrayList<String>();
+ IndexWord word = null;
+ Synset sense = null;
+ try {
+ word = dict.getIndexWord(POS.NOUN, s);
+ if (word != null) {
+ sense = word.getSense(1);
+ //Synset sense = word.getSense(1);
+
+ PointerTargetNodeList relatedListHypernyms = null;
+ PointerTargetNodeList relatedListHyponyms = null;
+ try {
+ relatedListHypernyms = PointerUtils.getInstance().getDirectHypernyms(sense);
+ }
+ catch (JWNLException e) {
+ // TODO Auto-generated catch block
+ e.printStackTrace();
+ }
+ try {
+ relatedListHyponyms = PointerUtils.getInstance().getDirectHyponyms(sense);
+ }
+ catch (JWNLException e) {
+ // TODO Auto-generated catch block
+ e.printStackTrace();
+ }
+
+ Iterator i = relatedListHypernyms.iterator();
+ while (i.hasNext()) {
+ PointerTargetNode related = (PointerTargetNode) i.next();
+ Synset s1 = related.getSynset();
+ String tmp = (s1.toString()).replace(s1.getGloss(), "");
+ tmp = tmp.replace(" -- ()]", "");
+ tmp = tmp.replaceAll("[0-9]", "");
+ tmp = tmp.replace("[Synset: [Offset: ", "");
+ tmp = tmp.replace("] [POS: noun] Words: ", "");
+ //its possible, that there is more than one word in a line from wordnet
+ String[] array_tmp = tmp.split(",");
+ for (String z : array_tmp) {
+ result.add(z.replace(" ", ""));
+ }
+ }
+
+ Iterator j = relatedListHyponyms.iterator();
+ while (j.hasNext()) {
+ PointerTargetNode related = (PointerTargetNode) j.next();
+ Synset s1 = related.getSynset();
+ String tmp = (s1.toString()).replace(s1.getGloss(), "");
+ tmp = tmp.replace(" -- ()]", "");
+ tmp = tmp.replaceAll("[0-9]", "");
+ tmp = tmp.replace("[Synset: [Offset: ", "");
+ tmp = tmp.replace("] [POS: noun] Words: ", "");
+ //its possible, that there is more than one word in a line from wordnet
+ String[] array_tmp = tmp.split(",");
+ for (String z : array_tmp) {
+ result.add(z.replace(" ", ""));
+ }
+ }
+ }
+ }
+ catch (JWNLException e) {
+ // TODO Auto-generated catch block
+ e.printStackTrace();
+ }
+
+
+ return result;
+ }
+
}
Modified: trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/LinguisticUtil.java
===================================================================
--- trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/LinguisticUtil.java 2013-09-05 13:59:47 UTC (rev 4091)
+++ trunk/components-core/src/main/java/org/dllearner/algorithms/isle/index/LinguisticUtil.java 2013-09-06 10:01:53 UTC (rev 4092)
@@ -83,6 +83,23 @@
}
/**
+ * Returns an array of the lemmas of the top {@code n} synonyms for the given word. Only synonyms for the POS in
+ * {@link #RELEVANT_POS} are returned.
+ *
+ * @param word the word to retrieve synonyms for
+ * @param n the number of senses to get lemmas for
+ * @return synonyms for the given word
+ */
+ public static String[] getTopSynonymsForWord(String word, int n) {
+ ArrayList<String> synonyms = new ArrayList<String>();
+
+ for (POS pos : RELEVANT_POS) {
+ synonyms.addAll(wn.getTopSynonyms(pos, word, n));
+ }
+ return synonyms.toArray(new String[synonyms.size()]);
+ }
+
+ /**
* Returns the normalized form of the given word. This method is only able to work with single words! If there is an
* error normalizing the given word, the word itself is returned.
*
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
