use of zemberek.morphology.analysis.SentenceWordAnalysis in project zemberek-nlp by ahmetaa.
the class QuestionClassifier method splitWords.
private String splitWords(String sentence) {
List<String> tokens = Splitter.on(" ").splitToList(sentence);
// assume first is label. Remove label from sentence for morphological analysis.
String label = tokens.get(0);
tokens = tokens.subList(1, tokens.size());
sentence = String.join(" ", tokens);
if (sentence.length() == 0) {
return sentence;
}
SentenceAnalysis analysis = morphology.analyzeAndDisambiguate(sentence);
List<String> res = new ArrayList<>();
// add label first.
res.add(label);
for (SentenceWordAnalysis e : analysis) {
SingleAnalysis best = e.getBestAnalysis();
if (best.isUnknown()) {
res.add(e.getWordAnalysis().getInput());
continue;
}
List<String> lemmas = best.getLemmas();
res.add(lemmas.get(lemmas.size() - 1));
}
return String.join(" ", res);
}
use of zemberek.morphology.analysis.SentenceWordAnalysis in project zemberek-nlp by ahmetaa.
the class Scripts method saveUnambiguous.
public static void saveUnambiguous(List<SentenceAnalysis> sentences, Path out) throws IOException {
try (PrintWriter pwMorph = new PrintWriter(out.toFile(), "utf-8")) {
for (SentenceAnalysis analysis : sentences) {
if (analysis.bestAnalysis().stream().anyMatch(SingleAnalysis::isUnknown)) {
continue;
}
pwMorph.format("S:%s%n", analysis.getSentence());
for (SentenceWordAnalysis sw : analysis) {
WordAnalysis wa = sw.getWordAnalysis();
pwMorph.println(wa.getInput());
SingleAnalysis best = sw.getBestAnalysis();
for (SingleAnalysis singleAnalysis : wa) {
boolean isBest = singleAnalysis.equals(best);
if (wa.analysisCount() == 1) {
pwMorph.println(singleAnalysis.formatLong());
} else {
pwMorph.format("%s%s%n", singleAnalysis.formatLong(), isBest ? "*" : "");
}
}
}
pwMorph.println();
}
}
}
use of zemberek.morphology.analysis.SentenceWordAnalysis in project zemberek-nlp by ahmetaa.
the class ClassificationExampleBase method splitWords.
protected String splitWords(String sentence) {
List<String> tokens = Splitter.on(" ").splitToList(sentence);
// assume first is label. Remove label from sentence for morphological analysis.
String label = tokens.get(0);
tokens = tokens.subList(1, tokens.size());
sentence = String.join(" ", tokens);
if (sentence.length() == 0) {
return sentence;
}
SentenceAnalysis analysis = morphology.analyzeAndDisambiguate(sentence);
List<String> res = new ArrayList<>();
// add label first.
res.add(label);
for (SentenceWordAnalysis e : analysis) {
SingleAnalysis best = e.getBestAnalysis();
String input = e.getWordAnalysis().getInput();
if (best.isUnknown()) {
res.add(input);
continue;
}
List<String> lemmas = best.getLemmas();
String l = lemmas.get(0);
if (l.length() < input.length()) {
res.add(l);
String substring = input.substring(l.length());
res.add("_" + substring);
} else {
res.add(l);
}
}
return String.join(" ", res);
}
use of zemberek.morphology.analysis.SentenceWordAnalysis in project zemberek-nlp by ahmetaa.
the class FindPOS method main.
public static void main(String[] args) {
TurkishMorphology morphology = TurkishMorphology.createWithDefaults();
String sentence = "Keşke yarın hava güzel olsa.";
Log.info("Sentence = " + sentence);
SentenceAnalysis analysis = morphology.analyzeAndDisambiguate(sentence);
for (SentenceWordAnalysis a : analysis) {
PrimaryPos primaryPos = a.getBestAnalysis().getPos();
Log.info("%s : %s ", a.getWordAnalysis().getInput(), primaryPos);
}
}
Aggregations