Package opennlp.tools.parser

Examples of opennlp.tools.parser.Parse

64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
      String tok = str.nextToken();
      tokens.add(tok);
      sb.append(tok).append(" ");
    }
    String text = sb.substring(0, sb.length() - 1);
    Parse p = new Parse(text, new Span(0, text.length()), AbstractBottomUpParser.INC_NODE, 0, 0);
    int start = 0;
    int i=0;
    for (Iterator<String> ti = tokens.iterator(); ti.hasNext();i++) {
      String tok = ti.next();
      p.insert(new Parse(text, new Span(start, start + tok.length()), AbstractBottomUpParser.TOK_NODE, 0,i));
      start += tok.length() + 1;
    }
    Parse[] parses;
    if (numParses == 1) {
      parses = new Parse[] { parser.parse(p)};
View Full Code Here
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
            sentenceNumber=0;
            document.clear();
            parses.clear();
          }
          else {
            Parse p = Parse.parseParse(line);
            parses.add(p);
            Mention[] extents = treebankLinker.getMentionFinder().getMentions(new DefaultParse(p,sentenceNumber));
            //construct new parses for mentions which don't have constituents.
            for (int ei=0,en=extents.length;ei<en;ei++) {
              //System.err.println("PennTreebankLiner.main: "+ei+" "+extents[ei]);

              if (extents[ei].getParse() == null) {
                //not sure how to get head index, but its not used at this point.
                Parse snp = new Parse(p.getText(),extents[ei].getSpan(),"NML",1.0,0);
                p.insert(snp);
                extents[ei].setParse(new DefaultParse(snp,sentenceNumber));
              }

            }
View Full Code Here
55
56
57
58
59
60
61
62
63
64
65
      parseMap = new HashMap<Parse, Integer>();
      for (int ei=0,en=entities.length;ei<en;ei++) {
        if (entities[ei].getNumMentions() > 1) {
          for (Iterator<MentionContext> mi = entities[ei].getMentions(); mi.hasNext();) {
            MentionContext mc = mi.next();
            Parse mentionParse = ((DefaultParse) mc.getParse()).getParse();
            parseMap.put(mentionParse,ei+1);
          }
        }
      }
    }
View Full Code Here
64
65
66
67
68
69
70
71
72
73
      }
    }

    public void show() {
      for (int pi=0,pn=parses.size();pi<pn;pi++) {
        Parse p = parses.get(pi);
        show(p);
        System.out.println();
      }
    }
View Full Code Here
84
85
86
87
88
89
90
91
92
93
94
95
        //System.out.print(p.hashCode()+"-"+parseMap.containsKey(p));
        System.out.print(" ");
      }
      Parse[] children = p.getChildren();
      for (int pi=0,pn=children.length;pi<pn;pi++) {
        Parse c = children[pi];
        Span s = c.getSpan();
        if (start < s.getStart()) {
          System.out.print(p.getText().substring(start, s.getStart()));
        }
        show(c);
        start = s.getEnd();
View Full Code Here
118
119
120
121
122
123
124
125
126
127
128
129
130
  public void testTokensAreCorrect() throws IOException {
   
    ObjectStream<Parse> samples =
        new ConstitParseSampleStream(ObjectStreamUtils.createObjectStream(getSample1()));
   
    Parse p = samples.read();
   
    Parse[] tagNodes = p.getTagNodes();
    String[] tokens = new String[tagNodes.length];
    for (int ti=0;ti<tagNodes.length;ti++){
      tokens[ti] = tagNodes[ti].getCoveredText();
    }
   
View Full Code Here
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
     
      List<String[]> sentences = sample.getTexts();
     
      for (String sentence[] : sentences) {
       
        Parse p = FullParseCorefEnhancerStream.createIncompleteParse(sentence);
        p.setType(AbstractBottomUpParser.TOP_NODE);
       
        Parse parseTokens[] = p.getChildren();
       
        // construct incomplete parse here ..
        String tags[] = posTagger.tag(sentence);
       
        for (int i = 0; i < parseTokens.length; i++) {
          p.insert(new Parse(p.getText(), parseTokens[i].getSpan(), tags[i], 1d, parseTokens[i].getHeadIndex()));
        }
       
        // insert tags into incomplete parse
        Span chunks[] = chunker.chunkAsSpans(sentence, tags);
       
        for (Span chunk : chunks) {
          if ("NP".equals(chunk.getType())) {
            p.insert(new Parse(p.getText(), new Span(0,0), chunk.getType(), 1d, p.getHeadIndex()));
          }
        }
       
        enhancedParses.add(p);
      }
View Full Code Here
121
122
123
124
125
126
127
128
129
130
131
132
133
134
  @Override
  protected void addParseEvents(List<Event> parseEvents, Parse[] chunks) {
    int ci = 0;
    while (ci < chunks.length) {
      //System.err.println("parserEventStream.addParseEvents: chunks="+Arrays.asList(chunks));
      Parse c = chunks[ci];
      Parse parent = c.getParent();
      if (parent != null) {
        String type = parent.getType();
        String outcome;
        if (firstChild(c, parent)) {
          outcome = AbstractBottomUpParser.START + type;
        }
        else {
View Full Code Here
41
42
43
44
45
46
47
48
49
50
51
52
  }

  private boolean containsPunct(Collection<Parse> puncts, String punct){
    if (puncts != null){
      for (Iterator<Parse> pi=puncts.iterator();pi.hasNext();) {
        Parse p = pi.next();
        if (p.getType().equals(punct)) {
          return true;
        }
      }
    }
    return false;
View Full Code Here
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
   * @return A set of contextual features about this attachment.
   */
  public String[] getContext(Parse[] constituents, int index, List<Parse> rightFrontier, int rfi) {
    List<String> features = new ArrayList<String>(100);
    int nodeDistance = rfi;
    Parse fn = rightFrontier.get(rfi);
    Parse fp = null;
    if (rfi+1 < rightFrontier.size()) {
      fp = rightFrontier.get(rfi+1);
    }
    Parse p_1 = null;
    if (rightFrontier.size() > 0) {
      p_1 = rightFrontier.get(0);
    }
    Parse p0 = constituents[index];
    Parse p1 = null;
    if (index+1 < constituents.length) {
      p1 = constituents[index+1];
    }

    Collection<Parse> punct1s = null;
    Collection<Parse> punct_1s = null;
    Collection<Parse> punct_1fs = null;
    punct_1fs = fn.getPreviousPunctuationSet();
    punct_1s=p0.getPreviousPunctuationSet();
    punct1s=p0.getNextPunctuationSet();

    String consfp = cons(fp,-3);
    String consf = cons(fn,-2);
    String consp_1 = cons(p_1,-1);
    String consp0 = cons(p0,0);
    String consp1 = cons(p1,1);

    String consbofp = consbo(fp,-3);
    String consbof = consbo(fn,-2);
    String consbop_1 = consbo(p_1,-1);
    String consbop0 = consbo(p0,0);
    String consbop1 = consbo(p1,1);

    Cons cfp = new Cons(consfp,consbofp,-3,true);
    Cons cf = new Cons(consf,consbof,-2,true);
    Cons c_1 = new Cons(consp_1,consbop_1,-1,true);
    Cons c0 = new Cons(consp0,consbop0,0,true);
    Cons c1 = new Cons(consp1,consbop1,1,true);

    //default
    features.add("default");

    //unigrams
    features.add(consfp);
    features.add(consbofp);
    features.add(consf);
    features.add(consbof);
    features.add(consp_1);
    features.add(consbop_1);
    features.add(consp0);
    features.add(consbop0);
    features.add(consp1);
    features.add(consbop1);

    //productions
    String prod = production(fn,false);
    //String punctProd = production(fn,true,punctSet);
    features.add("pn="+prod);
    features.add("pd="+prod+","+p0.getType());
    features.add("ps="+fn.getType()+"->"+fn.getType()+","+p0.getType());
    if (punct_1s != null) {
      StringBuilder punctBuf = new StringBuilder(5);
      for (Iterator<Parse> pi=punct_1s.iterator();pi.hasNext();) {
        Parse punct = pi.next();
        punctBuf.append(punct.getType()).append(",");
      }
      //features.add("ppd="+punctProd+","+punctBuf.toString()+p0.getType());
      //features.add("pps="+fn.getType()+"->"+fn.getType()+","+punctBuf.toString()+p0.getType());
    }

View Full Code Here
TOP

Related Classes of opennlp.tools.parser.Parse

  • com.tamingtext.opennlp.ChunkParserTest
  • com.tamingtext.qa.AnswerTypeContextGenerator
  • com.tamingtext.qa.AnswerTypeEventStream
  • com.tamingtext.qa.ChunkParser
  • com.tamingtext.qa.QuestionQParser
  • functionality.Parsing
  • net.sf.nlpshell.CorefParse
  • net.sf.nlpshell.Main
  • opennlp.tools.cmdline.coref.CoreferencerTool
  • opennlp.tools.cmdline.coref.CoreferencerTool$CorefParse
    • Copyright © 2018 www.massapicom. All rights reserved.
    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.