public List<String> annotatePOSpeech(List<String> tokenized)
throws IOException, TreeTaggerException {
// Point TT4J to the TreeTagger installation directory. The executable
// is expected
// in the "bin" subdirectory - in this example at
// "/opt/treetagger/bin/tree-tagger"
System.setProperty("treetagger.home", treeTaggerFolder);
TreeTaggerWrapper<String> tt = new TreeTaggerWrapper<String>();
final List<String> outputTreeTagger = new ArrayList<String>();
try {
tt.setModel(treeTaggerFolder + "lib/english.par:utf8");
tt.setHandler(
new TokenHandler<String>() {
public void token(String token, String pos, String lemma) {
outputTreeTagger.add(token + "\t" + pos + "\t" + lemma);
}
});
tt.process(tokenized);
} finally {
tt.destroy();
}
return outputTreeTagger;
}
public LemmatiseurHandler(String ch) {
termes = new ArrayList();
termesLem = new ArrayList();
termesPos = new ArrayList();
tt = new TreeTaggerWrapper<String>();
// Lemmatiseur
System.setProperty("treetagger.home", ch);
try {
tt.setModel("french.par:UTF8");
tt.setHandler(
new TokenHandler<String>() {
@Override
public void token(String token, String pos, String lemma) {
// System.out.println(token + "\t" + pos + "\t" + lemma);
termesLem.add(lemma);
termesPos.add(pos);
}
});
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
