public static IEnumerable <POS> Extract(string text, ref NLPCount count)
{
var result = new List <POS>();
workflow.analyze(text);
var sentences = workflow.getResultOfDocument(new jHanNanum.comm.Sentence(0, 0, false));
foreach (jHanNanum.comm.Sentence sentence in sentences)
{
count.SentenceCount++;
var eojeols = sentence.Eojeols;
foreach (var eojeol in eojeols)
{
count.WordsPhraseCount++;
for (int i = 0; i < eojeol.length; i++)
{
count.MorphemeCount++;
result.Add(new POS()
{
PosTag = eojeol.Tags[i],
Text = eojeol.Morphemes[i]
});
}
}
}
return(result);
}
public static IEnumerable <POS> Extract(string text, ref NLPCount count)
{
var segments = new List <POS>();
if (string.IsNullOrEmpty(text))
{
return(segments);
}
MeCabNode node = tagger.ParseToNode(text);
while (node != null)
{
if (node.CharType > 0)
{
if (node.Surface.Length <= 100)
{
segments.Add(new POS()
{
Text = node.Surface, PosTag = node.Feature.Split(',')[0]
});
}
}
node = node.Next;
}
return(segments);
}
public static IEnumerable <POS> Extract(string text, ref NLPCount count)
{
string normalized = TwitterKoreanProcessor.normalize(text).toString();
var tokenized = TwitterKoreanProcessor.tokenize(normalized);
var stemmed = TwitterKoreanProcessor.stem(tokenized);
return(stemmed.ToPosListFromTokens().Where(s => s.PosTag != "Punctuation" && s.PosTag != "Space"));
}
public static IEnumerable <POS> Extract(string text, ref NLPCount count)
{
var segment = new List <POS>();
if (string.IsNullOrEmpty(text))
{
return(segment);
}
var document = new Annotation(text);
pipeline.annotate(document);
var sentencesAnnotation = new SentencesAnnotation();
var tokensAnnotation = new TokensAnnotation();
var textAnnotation = new TextAnnotation();
var partOfSpeechAnnotation = new PartOfSpeechAnnotation();
java.util.ArrayList sentenceArrayList = (java.util.ArrayList)document.get(sentencesAnnotation.getClass());
var sentences = sentenceArrayList.toArray();
count.SentenceCount += sentences.Length;
for (int i = 0; i < sentences.Length; i++)
{
var sentence = (edu.stanford.nlp.util.CoreMap)sentences[i];
var tokenArray = ((java.util.ArrayList)sentence.get(tokensAnnotation.getClass()));
var tokens = tokenArray.toArray();
count.WordsPhraseCount += tokens.Length;
for (int j = 0; j < tokens.Length; j++)
{
var coreLabel = (edu.stanford.nlp.ling.CoreLabel)tokens[j];
string posTag = (string)coreLabel.get(partOfSpeechAnnotation.getClass());
string word = (string)coreLabel.get(textAnnotation.getClass());
if (word.Length <= 100)
{
segment.Add(new POS()
{
Text = word, PosTag = posTag
});
}
}
}
return(segment.ToList());
}
public static IEnumerable <POS> Extract(string text)
{
var count = new NLPCount();
return(Extract(text, ref count));
}
