Sha256: a5a299084adf35f0310d6bcd6f49be06bbe396b2419714901a8db9a9d78e5c0d
Contents?: true
Size: 1.57 KB
Versions: 45
Compression:
Stored size: 1.57 KB
Contents
package lingscope.algorithms;
import java.util.List;
import lingscope.structures.AnnotatedSentence;
/**
*
* @author shashank
*/
public abstract class Annotator {
protected String beginTag;
protected String interTag;
protected String otherTag;
public Annotator(String beginTag, String interTag, String otherTag) {
this.beginTag = beginTag;
this.interTag = interTag;
this.otherTag = otherTag;
}
public abstract void serializeAnnotator(String trainingFile, String modelFile);
public abstract AnnotatedSentence annotateSentence(String sentence, boolean isTokenized);
public abstract void loadAnnotator(String modelFile);
/**
* Checks if the given target phrase is negated in the given sentence. Only
* the first word of the target phrase is used
* @param sentence
* @param isTokenized
* @param targetPhrase
* @return
*/
public boolean isTargetNegated(String sentence, boolean isTokenized, String targetPhrase) {
AnnotatedSentence annotatedSentence = annotateSentence(sentence, isTokenized);
String[] targetPhraseWords = targetPhrase.split("\\s+");
List<String> words = annotatedSentence.getWords();
List<Boolean> areNegated = annotatedSentence.getIsAnnotatedTags();
int index = 0;
for (String word : words) {
if (targetPhraseWords[0].equalsIgnoreCase(word)) {
return areNegated.get(index);
}
++index;
}
System.err.println("Phrase not found: " + targetPhrase);
return false;
}
}
Version data entries
45 entries across 45 versions & 1 rubygems