Examples of edu.stanford.nlp.ling.HasWord.word()

edu.stanford.nlp.ling.HasWord.word()
Return the word value of the label (or null if none). @return String the word value for the label

        } else if (seenBoundary && !delimFollowers.contains(token.word())) {
          nextSentCarryover.add(token);
          break;
        }


        if ( ! (wsPattern.matcher(token.word()).matches() ||
                token.word().equals(PTBLexer.NEWLINE_TOKEN))) {
          nextSent.add(token);
        }


        // If there are no words that can follow a sentence delimiter,

View Full Code Here

          nextSentCarryover.add(token);
          break;
        }


        if ( ! (wsPattern.matcher(token.word()).matches() ||
                token.word().equals(PTBLexer.NEWLINE_TOKEN))) {
          nextSent.add(token);
        }


        // If there are no words that can follow a sentence delimiter,
        // then there are two cases.  In one case is we already have a

View Full Code Here

    if (input.get(inputSize - 1).word().equals("\"")) {
      // alternate from the end
      begin = false;
      for (int i = inputSize - 1; i >= 0; i--) {
        HasWord hw = input.get(i);
        String tok = hw.word();
        if (tok.equals("\"")) {
          if (begin) {
            hw.setWord("``");
            begin = false;
          } else {

View Full Code Here

    } else {
      // alternate from the beginning
      begin = true;
      for (int i = 0; i < inputSize; i++) {
        HasWord hw = input.get(i);
        String tok = hw.word();
        if (tok.equals("\"")) {
          if (begin) {
            hw.setWord("``");
            begin = false;
          } else {

View Full Code Here


  @SuppressWarnings("OverlyStrongTypeCast")
  private String getString(IN o) {
    if (o instanceof HasWord) {
      HasWord h = (HasWord) o;
      return h.word();
    } else if (o instanceof String) {
      return (String) o;
    } else if (o instanceof CoreMap) {
      return ((CoreMap) o).get(CoreAnnotations.TextAnnotation.class);
    } else {

View Full Code Here

      taggedWordList[i] = new ArrayList<IntTaggedWord>(numTags);
      HasWord wordObject = wordList.get(i);
      if (wordObject instanceof CoreLabel) {
        originalLabels[i] = (CoreLabel) wordObject;
      }
      String wordStr = wordObject.word();


      //Word context (e.g., morphosyntactic info)
      String wordContextStr = null;
      if(wordObject instanceof HasContext) {
        wordContextStr = ((HasContext) wordObject).originalText();

View Full Code Here


      while (tokenizer.hasNext()) {


        HasWord token = tokenizer.next();
        if (splitTag != null) {
          String[] toks = splitTag.apply(token.word());
          token.setWord(toks[0]);
          if(toks.length == 2 && token instanceof HasTag) {
            //wsg2011: Some of the underlying tokenizers return old
            //JavaNLP labels.  We could convert to CoreLabel here, but
            //we choose a conservative implementation....

View Full Code Here

            //we choose a conservative implementation....
            ((HasTag) token).setTag(toks[1]);
          }
        }


        if (sentDelims.contains(token.word())) {
          seenBoundary = true;
        } else if (seenBoundary && !delimFollowers.contains(token.word())) {
          nextSentCarryover.add(token);
          break;
        }

View Full Code Here

          }
        }


        if (sentDelims.contains(token.word())) {
          seenBoundary = true;
        } else if (seenBoundary && !delimFollowers.contains(token.word())) {
          nextSentCarryover.add(token);
          break;
        }


        if ( ! (token.word().matches("\\s+") //||

View Full Code Here

        } else if (seenBoundary && !delimFollowers.contains(token.word())) {
          nextSentCarryover.add(token);
          break;
        }


        if ( ! (token.word().matches("\\s+") //|| 
                /*token.word().equals(PTBLexer.NEWLINE_TOKEN)*/)) {
          nextSent.add(token);
        }


        // If there are no words that can follow a sentence delimiter,

View Full Code Here

0 1 2

TOP

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.