Package com.digitalpebble.classification

Examples of com.digitalpebble.classification.Document

223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
        // get an iterator on the Corpus
        // and retrieve the documents one by one
        Iterator<Document> docIterator = corpus.iterator();
        while (docIterator.hasNext()) {
            Document doc = docIterator.next();
            int label = doc.getLabel();
            // get a vector from the document
            // need a metric (e.g. relative frequency / binary)
            // and a lexicon
            // the vector is represented as a string directly
            Vector vector = null;
            if (attributeMapping == null)
                vector = doc.getFeatureVector(lexicon);
            else
                vector = doc.getFeatureVector(lexicon, attributeMapping);

            StringBuffer buffer = new StringBuffer();

            buffer.append(lexicon.getLabel(label));
View Full Code Here
140
141
142
143
144
145
146
147
148
149
150
      // lower case
      for (int i=0;i<tokens.length;i++){
        tokens[i]=tokens[i].toLowerCase();
      }     
     
      Document doc = null;
      if (operator instanceof Learner)
      doc = ((Learner)operator).createDocument(tokens,label);
      else
        doc = ((TextClassifier)operator).createDocument(tokens);
      corpusList.add(doc);
View Full Code Here
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
        "title" });
    fields[1] = new Field("abstract", new String[] { "abstract" });
    fields[2] = new Field("content", new String[] { "This", "is", "the",
        "content", "this", "will", "have", "a", "large", "value" });
    learner.setMethod(Parameters.WeightingMethod.TFIDF);
    Document doc = learner.createDocument(fields, "large");

    Field[] fields2 = new Field[2];
    fields2[0] = new Field("title", new String[] { "This", "is", "not",
        "a", "title" });
    // fields2[1] = new Field("abstract", new String[]{"abstract"});
    fields2[1] = new Field("content", new String[] { "This", "is", "the",
        "content", "this", "will", "have", "a", "small", "value" });
    learner.setMethod(Parameters.WeightingMethod.TFIDF);
    Document doc2 = learner.createDocument(fields2, "small");

    // try putting the same field several times
    Field[] fields3 = new Field[3];
    fields3[0] = new Field("title", new String[] { "This", "is", "not",
        "a", "title" });
    // fields2[1] = new Field("abstract", new String[]{"abstract"});
    fields3[1] = new Field("content", new String[] { "This", "is", "the",
        "content", "this", "will", "have", "a", "small", "value" });
    fields3[2] = new Field("title", new String[] { "some", "different",
        "content" });
    learner.setMethod(Parameters.WeightingMethod.TFIDF);
    Document doc3 = learner.createDocument(fields3, "small");

    RAMTrainingCorpus corpus = new RAMTrainingCorpus();
    corpus.add(doc);
    corpus.add(doc2);
    learner.learn(corpus);
View Full Code Here
75
76
77
78
79
80
81
82
83
84
85
86
  public void testCustomWeightingScheme() throws Exception {
    Field[] fields = new Field[1];
    fields[0] = new Field("keywords", new String[] { "test","keywords"});
    learner.setMethod(Parameters.WeightingMethod.FREQUENCY);
    learner.getLexicon().setMethod(Parameters.WeightingMethod.BOOLEAN, "keywords");
    Document doc = learner.createDocument(fields, "large");
    Vector vector = doc.getFeatureVector(learner.getLexicon());
   
    // check that the values for the field keywords are boolean
    int[] indices = vector.getIndices();
    double[] values = vector.getValues();
   
View Full Code Here
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
    fields[0] = new Field("title", new String[] { "This", "is", "a",
        "title" });
    fields[1] = new Field("abstract", new String[] { "abstract" });
    fields[2] = new Field("content", new String[] { "This", "is", "the",
        "content", "this", "will", "have", "a", "large", "value" });
    Document doc = learner.createDocument(fields, "large");

    String[] simplecontent = new String[] { "This", "is", "the", "content",
        "this", "will", "have", "a", "small", "value" };
    Document doc2 = learner.createDocument(simplecontent, "small");

    // com.digitalpebble.classification.TrainingCorpus tc =
    // learner.getFileTrainingCorpus();
    com.digitalpebble.classification.TrainingCorpus tc = new RAMTrainingCorpus();
    tc.addDocument(doc);
View Full Code Here
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
    RAMTrainingCorpus corpus = new RAMTrainingCorpus();

    learner.setMethod(method);

    for (String[] content : docs) {
      Document doc = learner.createDocument(content);
      corpus.add(doc);
    }

    Iterator<Document> corpusIter = corpus.iterator();

    Map<Integer, String> invertedIndex = learner.getLexicon()
        .getInvertedIndex();

    List<Map> expectedset = references.get(method);

    // check that we have the same number of docs in the corpus
    // and in the ref

    assertEquals(expectedset.size(), corpus.size());

    for (Map<String, Double> ref : expectedset) {
      Document doc = corpusIter.next();
      Vector vector = doc.getFeatureVector(learner.getLexicon());
      // now let's compare what we wanted to have with the content of the
      // vector
      int[] indices = vector.getIndices();
      double[] values = vector.getValues();
View Full Code Here
TOP

Related Classes of com.digitalpebble.classification.Document

  • com.digitalpebble.classification.libsvm.Utils
  • com.digitalpebble.classification.test.classifyDoc
  • com.digitalpebble.classification.test.TestMultiFieldDocs
  • com.digitalpebble.classification.test.TestTrainingCorpus
  • com.digitalpebble.classification.test.TestWeightingSchemes
  • com.digitalpebble.classification.util.ModelUtils
  • com.digitalpebble.classification.util.scorers.chiSquareAttributeScorer
  • com.digitalpebble.classification.util.scorers.logLikelihoodAttributeScorer
    • 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.