Package org.apache.lucene.index

Examples of org.apache.lucene.index.IndexReader.docFreq()

34
35
36
37
38
39
40
41
42
43
44
      Term t = textEnum.term();
      if("txt".equals(t.field())) {
        String text = t.text();
        if(TermSets.getClosedClass().contains(text)) continue;
        if(!text.matches(".*[A-Za-z].*")) continue;
        int docFreq = ir.docFreq(t);
        if(docFreq > 1) {
          docFreqs.put(text, ir.docFreq(t));
        }
      }     
    }
View Full Code Here
36
37
38
39
40
41
42
43
44
45
46
        String text = t.text();
        if(TermSets.getClosedClass().contains(text)) continue;
        if(!text.matches(".*[A-Za-z].*")) continue;
        int docFreq = ir.docFreq(t);
        if(docFreq > 1) {
          docFreqs.put(text, ir.docFreq(t));
        }
      }     
    }
    List<String> dfl = StringTools.getSortedList(docFreqs);
    Map<String,Integer> termMap = new HashMap<String,Integer>();
View Full Code Here
37
38
39
40
41
42
43
44
45
46
47
      Term t = textEnum.term();
      if("txt".equals(t.field())) {
        String text = t.text();
        if(TermSets.getClosedClass().contains(text)) continue;
        if(!text.matches(".*[A-Za-z].*")) continue;
        int docFreq = ir.docFreq(t);
        if(docFreq > 0) {
          docFreqs.put(text, ir.docFreq(t));
        }
      }     
    }
View Full Code Here
39
40
41
42
43
44
45
46
47
48
49
        String text = t.text();
        if(TermSets.getClosedClass().contains(text)) continue;
        if(!text.matches(".*[A-Za-z].*")) continue;
        int docFreq = ir.docFreq(t);
        if(docFreq > 0) {
          docFreqs.put(text, ir.docFreq(t));
        }
      }     
    }
    List<String> dfl = StringTools.getSortedList(docFreqs);
    int maxSize = 20000;
View Full Code Here
158
159
160
161
162
163
164
165
166
167
168
      Term t = textEnum.term();
      if("txt".equals(t.field())) {
        String text = t.text();
        if(TermSets.getClosedClass().contains(text)) continue;
        if(!text.matches(".*[A-Za-z].*")) continue;
        int docFreq = ir.docFreq(t);
        if(docFreq > 1) {
          docFreqs.put(text, ir.docFreq(t));
        }
      }     
    }
View Full Code Here
160
161
162
163
164
165
166
167
168
169
170
        String text = t.text();
        if(TermSets.getClosedClass().contains(text)) continue;
        if(!text.matches(".*[A-Za-z].*")) continue;
        int docFreq = ir.docFreq(t);
        if(docFreq > 1) {
          docFreqs.put(text, ir.docFreq(t));
        }
      }     
    }
    List<String> dfl = StringTools.getSortedList(docFreqs);
   
View Full Code Here
322
323
324
325
326
327
328
329
330
331
332
        for(int j=0;j<lm.getColumnDimension();j++) {
          squareLength += Math.pow(lm.get(i, j), 2);
        }
        double distance = Math.sqrt(squareLength);
        String term = dfl.get(i);
        int docFreq = ir.docFreq(new Term("txt", term));
        double score = Math.log(docFreq) * distance;
        if(score > 1.0) distances.put(dfl.get(i), score);
      }
      out.close();
     
View Full Code Here
43
44
45
46
47
48
49
50
51
52
53
      Term t = textEnum.term();
      if("txt".equals(t.field())) {
        String text = t.text();
        if(TermSets.getClosedClass().contains(text)) continue;
        if(!text.matches(".*[A-Za-z].*")) continue;
        int docFreq = ir.docFreq(t);
        if(docFreq > 1) {
          docFreqs.put(text, ir.docFreq(t));
        }
      }     
    }
View Full Code Here
45
46
47
48
49
50
51
52
53
54
55
        String text = t.text();
        if(TermSets.getClosedClass().contains(text)) continue;
        if(!text.matches(".*[A-Za-z].*")) continue;
        int docFreq = ir.docFreq(t);
        if(docFreq > 1) {
          docFreqs.put(text, ir.docFreq(t));
        }
      }     
    }
    List<String> dfl = StringTools.getSortedList(docFreqs);

View Full Code Here
209
210
211
212
213
214
215
216
217
218
219
    //evd.getD().print(6, 2);

    Map<String,Double> termTots = new HashMap<String,Double>();
    for(int i=0;i<evr.length;i++) {
      for(String term : vectors.get(i).keySet()) {
        if(ir.docFreq(new Term("txt", term)) < 10) continue;
        if(!termTots.containsKey(term)) termTots.put(term, 0.0);
        termTots.put(term, termTots.get(term) + (vectors.get(i).get(term)));
      }
    }
    //System.out.println("termTots: " + termTots.size());
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.