lkrsnik/list
1
0
Fork 0
Code Issues 1 Pull Requests Releases Wiki Activity

Added GOS normalized words analysis in letter extraction + Fixing normalized words bugs with anonymous names in GOS (when extracting data with in collocability)

Browse Source
This commit is contained in:
Luka 2019-07-31 17:53:32 +02:00
parent e103bfa38d
commit edcd8062bc
6 changed files with 80 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/main/java/data/MultipleHMKeys2.java
View File

@ -27,6 +27,10 @@ public final class MultipleHMKeys2 implements MultipleHMKeys {
String[] splitedK1 = k1.split("\\s+");
String[] splitedK2 = k2.split("\\s+");
for(int i = 0; i < splitedK1.length; i ++){
// in GOS words and normalized words may not both have specific word due to anon
if(!(i < splitedK2.length)){
continue;
}
MultipleHMKeys search = new MultipleHMKeys2(splitedK1[i], splitedK2[i]);
r.add(search);
}

4
src/main/java/data/MultipleHMKeys3.java
View File

@ -33,6 +33,10 @@ public final class MultipleHMKeys3 implements MultipleHMKeys {
String[] splitedK2 = k2.split("\\s+");
String[] splitedK3 = k3.split("\\s+");
for(int i = 0; i < splitedK1.length; i ++){
// in GOS words and normalized words may not both have specific word due to anon
if(!(i < splitedK2.length && i < splitedK3.length)){
continue;
}
MultipleHMKeys search = new MultipleHMKeys3(splitedK1[i], splitedK2[i], splitedK3[i]);
r.add(search);
}

4
src/main/java/data/MultipleHMKeys4.java
View File

@ -39,6 +39,10 @@ public final class MultipleHMKeys4 implements MultipleHMKeys {
String[] splitedK3 = k3.split("\\s+");
String[] splitedK4 = k4.split("\\s+");
for(int i = 0; i < splitedK1.length; i ++){
// in GOS words and normalized words may not both have specific word due to anon
if(!(i < splitedK2.length && i < splitedK3.length && i < splitedK4.length)){
continue;
}
MultipleHMKeys search = new MultipleHMKeys4(splitedK1[i], splitedK2[i], splitedK3[i], splitedK4[i]);
r.add(search);
}

4
src/main/java/data/MultipleHMKeys5.java
View File

@ -45,6 +45,10 @@ public final class MultipleHMKeys5 implements MultipleHMKeys {
String[] splitedK4 = k4.split("\\s+");
String[] splitedK5 = k5.split("\\s+");
for(int i = 0; i < splitedK1.length; i ++){
// in GOS words and normalized words may not both have specific word due to anon
if(!(i < splitedK2.length && i < splitedK3.length && i < splitedK4.length && i < splitedK5.length)){
continue;
}
MultipleHMKeys search = new MultipleHMKeys5(splitedK1[i], splitedK2[i], splitedK3[i], splitedK4[i], splitedK5[i]);
r.add(search);
}

9
src/main/java/data/Word.java
View File

@ -108,7 +108,14 @@ public interface Word {
if (cvv) {
returnValue = (calculateFor == CalculateFor.WORD || calculateFor == CalculateFor.LOWERCASE_WORD) ? getCVVWord(cf) : getCVVLemma(cf);
} else {
returnValue = (calculateFor == CalculateFor.WORD || calculateFor == CalculateFor.LOWERCASE_WORD) ? getWord(cf) : getLemma(cf);
if (calculateFor == CalculateFor.WORD || calculateFor == CalculateFor.LOWERCASE_WORD){
returnValue = getWord(cf);
} else if (calculateFor == CalculateFor.LEMMA) {
returnValue = getLemma(cf);
} else if (calculateFor == CalculateFor.NORMALIZED_WORD){
returnValue = getNormalizedWord(cf);
}
// returnValue = (calculateFor == CalculateFor.WORD || calculateFor == CalculateFor.LOWERCASE_WORD) ? getWord(cf) : getLemma(cf);
}
return returnValue;

69
src/main/java/gui/CharacterAnalysisTab.java
View File

@ -161,6 +161,7 @@ public class CharacterAnalysisTab {
private HashMap<String, HashSet<String>> solarFiltersMap;
private HostServices hostService;
private ListChangeListener<String> taxonomyListener;
private ChangeListener<String> calculateForListener;
private ChangeListener<Boolean> msdListener;
private ChangeListener<Boolean> minimalOccurrencesListener;
private ChangeListener<Boolean> minimalTaxonomyListener;
@ -169,6 +170,9 @@ public class CharacterAnalysisTab {
private static final String [] N_GRAM_COMPUTE_FOR_LETTERS_ARRAY = {"calculateFor.WORD", "calculateFor.LOWERCASE_WORD", "calculateFor.LEMMA"};
private static final ArrayList<String> N_GRAM_COMPUTE_FOR_LETTERS = new ArrayList<>(Arrays.asList(N_GRAM_COMPUTE_FOR_LETTERS_ARRAY));
private static final String [] N_GRAM_COMPUTE_FOR_WORDS_GOS_ARRAY = {"calculateFor.WORD", "calculateFor.LOWERCASE_WORD", "calculateFor.LEMMA", "calculateFor.NORMALIZED_WORD"};
private static final ArrayList<String> N_GRAM_COMPUTE_FOR_WORDS_GOS = new ArrayList<>(Arrays.asList(N_GRAM_COMPUTE_FOR_WORDS_GOS_ARRAY));
private static final String [] TAXONOMY_SET_OPERATION_ARRAY = {"taxonomySetOperation.UNION", "taxonomySetOperation.INTERSECTION"};
private static final ArrayList<String> TAXONOMY_SET_OPERATION = new ArrayList<>(Arrays.asList(TAXONOMY_SET_OPERATION_ARRAY));
@ -184,14 +188,46 @@ public class CharacterAnalysisTab {
currentMode = MODE.LETTER;
toggleMode(currentMode);
calculateForCB.valueProperty().addListener((observable, oldValue, newValue) -> {
if(newValue == null){
newValue = I18N.getTranslatedValue(oldValue, N_GRAM_COMPUTE_FOR_LETTERS);
calculateForCB.getSelectionModel().select(newValue);
if (calculateForListener != null){
calculateForCB.valueProperty().removeListener(calculateForListener);
}
// calculateForCB
calculateForListener = new ChangeListener<String>() {
boolean ignoreCode = false;
@Override
public void changed(ObservableValue<? extends String> observable, String oldValue, String newValue) {
if (ignoreCode) {
return;
}
boolean languageChanged = newValue == null;
if (languageChanged) {
ignoreCode = true;
if (corpus.getCorpusType() == CorpusType.GOS) {
newValue = I18N.getTranslatedValue(oldValue, N_GRAM_COMPUTE_FOR_WORDS_GOS);
calculateForCB.getSelectionModel().select(newValue);
} else {
newValue = I18N.getTranslatedValue(oldValue, N_GRAM_COMPUTE_FOR_LETTERS);
calculateForCB.getSelectionModel().select(newValue);
}
ignoreCode = false;
}
calculateFor = CalculateFor.factory(newValue);
logger.info("calculateForCB:", calculateFor.toString());
}
calculateFor = CalculateFor.factory(newValue);
logger.info("calculateForCB:", calculateFor.toString());
});
};
calculateForCB.valueProperty().addListener(calculateForListener);
// calculateForCB.valueProperty().addListener((observable, oldValue, newValue) -> {
// if(newValue == null){
// newValue = I18N.getTranslatedValue(oldValue, N_GRAM_COMPUTE_FOR_LETTERS);
// calculateForCB.getSelectionModel().select(newValue);
// }
// calculateFor = CalculateFor.factory(newValue);
// logger.info("calculateForCB:", calculateFor.toString());
// });
calculateForCB.getSelectionModel().select(0);
@ -497,12 +533,19 @@ public class CharacterAnalysisTab {
stringLengthTF.setText(String.valueOf(stringLength));
}
// if calculateFor was selected for something other than a word or a lemma -> reset
if (!(calculateFor == CalculateFor.WORD || calculateFor == CalculateFor.LEMMA || calculateFor == CalculateFor.LOWERCASE_WORD)) {
// if the user selected something else before selecting ngram for letters, reset that choice
calculateFor = CalculateFor.WORD;
calculateForCB.getSelectionModel().select(0);
// // if calculateFor was selected for something other than a word or a lemma -> reset
// if (!(calculateFor == CalculateFor.WORD || calculateFor == CalculateFor.LEMMA || calculateFor == CalculateFor.LOWERCASE_WORD)) {
// // if the user selected something else before selecting ngram for letters, reset that choice
// calculateFor = CalculateFor.WORD;
//
// calculateForCB.getSelectionModel().select(0);
// }
if (corpus.getCorpusType() == CorpusType.GOS) {
calculateForCB.itemsProperty().unbind();
calculateForCB.itemsProperty().bind(I18N.createObjectBinding(N_GRAM_COMPUTE_FOR_WORDS_GOS));
} else {
calculateForCB.itemsProperty().unbind();
calculateForCB.itemsProperty().bind(I18N.createObjectBinding(N_GRAM_COMPUTE_FOR_LETTERS));
}
}