Package org.carrot2.core

Examples of org.carrot2.core.Cluster$DocumentRefid

73
74
75
76
77
78
79
80
81
82
83
        processingAttributes.put(AttributeUtils.getKey(ByFieldClusteringAlgorithm.class,
            "fieldName"), FIELD);
        final List<Cluster> clusters = cluster(documents).getClusters();
        if (unclustered.length > 0)
        {
            final Cluster otherTopics = new Cluster("Other Topics", unclustered);
            otherTopics.setOtherTopics(true);
            expectedClusters.add(otherTopics);
        }
        assertThatClusters(clusters).isEquivalentTo(expectedClusters);
    }
View Full Code Here
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
                }
            }

            for (int i = 0; i < rawClusters.size(); i++)
            {
                final Cluster cluster = new Cluster();

                final IntArrayList rawCluster = rawClusters.get(i);
                if (rawCluster.size() > 1)
                {
                    cluster.addPhrases(getLabels(rawCluster,
                        vsmContext.termDocumentMatrix, rowToStemIndex,
                        preprocessingContext.allStems.mostFrequentOriginalWordIndex,
                        preprocessingContext.allWords.image));
                    for (int j = 0; j < rawCluster.size(); j++)
                    {
                        cluster.addDocuments(documents.get(rawCluster.get(j)));
                    }
                    clusters.add(cluster);
                }
            }
        }
View Full Code Here
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
            final int [] clusterLabelIndex = lingoContext.clusterLabelFeatureIndex;
            final BitSet [] clusterDocuments = lingoContext.clusterDocuments;
            final double [] clusterLabelScore = lingoContext.clusterLabelScore;
            for (int i = 0; i < clusterLabelIndex.length; i++)
            {
                final Cluster cluster = new Cluster();

                final int labelFeature = clusterLabelIndex[i];
                if (labelFeature < 0)
                {
                    // Cluster removed during merging
                    continue;
                }

                // Add label and score
                cluster.addPhrases(labelFormatter.format(context, labelFeature));
                cluster.setAttribute(Cluster.SCORE, clusterLabelScore[i]);

                // Add documents
                final BitSet bs = clusterDocuments[i];
                for (int bit = bs.nextSetBit(0); bit >= 0; bit = bs.nextSetBit(bit + 1))
                {
                    cluster.addDocuments(documents.get(bit));
                }

                // Add cluster
                clusters.add(cluster);
            }
View Full Code Here
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
            final Set<Document> partitionDocuments = documentsByPartition.get(partition);
            final int partitionDocumentsCount = partitionDocuments.size();
            double partitionFMeasure = 0;
            double partitionPrecision = 0;
            double partitionRecall = 0;
            Cluster bestFMeasureCluster = null;

            for (Cluster cluster : clusters)
            {
                final List<Document> clusterDocuments = cluster.getAllDocuments();
                if (cluster.isOtherTopics() || clusterDocuments.size() == 0)
                {
                    continue;
                }

                final Set<Document> commonDocuments = Sets.newHashSet(partitionDocuments);
                commonDocuments.retainAll(clusterDocuments);

                final double precision = commonDocuments.size()
                    / (double) clusterDocuments.size();
                final double recall = commonDocuments.size()
                    / (double) partitionDocumentsCount;
                final double fMeasure = MathUtils.harmonicMean(precision, recall);

                if (fMeasure > partitionFMeasure)
                {
                    partitionFMeasure = fMeasure;
                    partitionPrecision = precision;
                    partitionRecall = recall;
                    bestFMeasureCluster = cluster;
                }
            }

            recallSum += partitionRecall * partitionDocumentsCount;
            precisionSum += partitionPrecision * partitionDocumentsCount;
            fMeasureSum += partitionFMeasure * partitionDocumentsCount;
            partitionDocumentsCountSum += partitionDocumentsCount;

            recallByPartition.put(partition, partitionRecall);
            precisionByPartition.put(partition, partitionPrecision);
            fMeasureByPartition.put(partition, partitionFMeasure);
            if (bestFMeasureCluster != null)
            {
                bestFMeasureCluster.setAttribute(BEST_F_MEASURE_PARTITION, partition);
            }
        }

        // Dividing by partitionDocumentsCountSum rather than by the number of documents
        // because partitionDocumentsCountSum can be larger than the number of documents
View Full Code Here
24
25
26
27
28
29
30
public class NormalizedMutualInformationMetricTest extends IdealPartitioningBasedMetricTest
{
    @Test
    public void testEmptyCluster()
    {
        check(null, new Cluster());
    }
View Full Code Here
30
31
32
33
34
35
36
    }

    @Test
    public void testTrivialCluster()
    {
        check(0.0, new Cluster("test", documentWithPartitions("test")));
    }
View Full Code Here
78
79
80
81
82
83
84
85
86
    }

    @Test
    public void testAllDocumentsInOtherTopics()
    {
        final Cluster otherTopics = clusterWithPartitions("t1", "t2", "t3");
        otherTopics.setOtherTopics(true);
        check(0.0, otherTopics);
    }
View Full Code Here
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
        Iterator<Cluster> expectedIt = expected.iterator();
        Iterator<Cluster> actualIt = actual.iterator();
        while (expectedIt.hasNext() && actualIt.hasNext())
        {
            final Cluster actualCluster = actualIt.next();
            final Cluster expectedCluster = expectedIt.next();
            for (ClusterPairCheck clusterPairAssert : clusterAsserts)
            {
                clusterPairAssert.check(actualCluster, expectedCluster);
            }

            assertThatClusters(actualCluster.getSubclusters()).as(
                description() + ": subclusters of \"" + actualCluster.getLabel() + "\"")
                .passRecursively(expectedCluster.getSubclusters(), clusterAsserts);
        }

        return this;
    }
View Full Code Here
238
239
240
241
242
243
244
245
246
247
248
249
        final IAdapterManager mgr = Platform.getAdapterManager();

        final BitSet newSelection = new BitSet();
        for (Object selected : ss.toArray())
        {
            final Cluster c = (Cluster) mgr.getAdapter(selected, Cluster.class);            
            if (c == null) continue;
            newSelection.set(c.getId());
        }
       
        return newSelection;
    }
View Full Code Here
35
36
37
38
39
40
41
42
43
44
45
46
47
        final String label;
        final int documentCount;

        if (element instanceof Cluster)
        {
            final Cluster cluster = (Cluster) element;
            label = cluster.getLabel();
            documentCount = cluster.size();
        }
        else
        {
            return "<unknown node: " + element + ">";
        }
View Full Code Here
TOP

Related Classes of org.carrot2.core.Cluster$DocumentRefid

  • com.carrotsearch.hppc.BitSet
  • com.carrotsearch.hppc.IntArrayList
  • com.carrotsearch.hppc.IntIntOpenHashMap
  • com.carrotsearch.hppc.IntStack
  • com.tamingtext.carrot2.Carrot2ExampleTest
  • org.apache.http.message.BasicNameValuePair
  • org.apache.lucene.search.IndexSearcher
  • org.apache.mahout.math.matrix.DoubleMatrix2D
  • org.carrot2.cli.batch.BatchApp
  • org.carrot2.clustering.kmeans.BisectingKMeansClusteringAlgorithm
    • 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.