2 files changed, 133 insertions, 0 deletions
diff --git a/test/de/lmu/ifi/dbs/elki/evaluation/TestComputeROC.java b/test/de/lmu/ifi/dbs/elki/evaluation/TestComputeROC.java
new file mode 100644
index 00000000..46adb1bb
--- /dev/null
+++ b/test/de/lmu/ifi/dbs/elki/evaluation/TestComputeROC.java
@@ -0,0 +1,55 @@
+package de.lmu.ifi.dbs.elki.evaluation;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import junit.framework.Assert;
+
+import org.junit.Test;
+
+import de.lmu.ifi.dbs.elki.JUnit4Test;
+import de.lmu.ifi.dbs.elki.database.ids.DBID;
+import de.lmu.ifi.dbs.elki.database.ids.DBIDUtil;
+import de.lmu.ifi.dbs.elki.database.ids.HashSetModifiableDBIDs;
+import de.lmu.ifi.dbs.elki.evaluation.roc.ROC;
+import de.lmu.ifi.dbs.elki.utilities.pairs.DoubleDoublePair;
+import de.lmu.ifi.dbs.elki.utilities.pairs.Pair;
+
+/**
+ * Test to validate ROC curve computation.
+ * 
+ * @author Erich Schubert
+ *
+ */
+public class TestComputeROC implements JUnit4Test {
+  /**
+   * Test ROC curve generation, including curve simplification
+   */
+  @Test
+  public void testROCCurve() {
+    HashSetModifiableDBIDs positive = DBIDUtil.newHashSet();
+    positive.add(DBIDUtil.importInteger(1));
+    positive.add(DBIDUtil.importInteger(2));
+    positive.add(DBIDUtil.importInteger(3));
+    positive.add(DBIDUtil.importInteger(4));
+    positive.add(DBIDUtil.importInteger(5));
+    
+    ArrayList<Pair<Double, DBID>> distances = new ArrayList<Pair<Double, DBID>>();
+    distances.add(new Pair<Double, DBID>(0.0, DBIDUtil.importInteger(1)));
+    distances.add(new Pair<Double, DBID>(1.0, DBIDUtil.importInteger(2)));
+    distances.add(new Pair<Double, DBID>(2.0, DBIDUtil.importInteger(6)));
+    distances.add(new Pair<Double, DBID>(3.0, DBIDUtil.importInteger(7)));
+    distances.add(new Pair<Double, DBID>(3.0, DBIDUtil.importInteger(3)));
+    distances.add(new Pair<Double, DBID>(4.0, DBIDUtil.importInteger(8)));
+    distances.add(new Pair<Double, DBID>(4.0, DBIDUtil.importInteger(4)));
+    distances.add(new Pair<Double, DBID>(5.0, DBIDUtil.importInteger(9)));
+    distances.add(new Pair<Double, DBID>(6.0, DBIDUtil.importInteger(5)));
+    
+    List<DoubleDoublePair> roccurve = ROC.materializeROC(9, positive, distances.iterator());
+    // System.out.println(roccurve);
+    Assert.assertEquals("ROC curve too complex", 6, roccurve.size());
+    
+    double auc = ROC.computeAUC(roccurve);
+    Assert.assertEquals("ROC AUC not right.", 0.6, auc, 0.0001);
+  }
+}
diff --git a/test/de/lmu/ifi/dbs/elki/evaluation/TestPairCountingFMeasure.java b/test/de/lmu/ifi/dbs/elki/evaluation/TestPairCountingFMeasure.java
new file mode 100644
index 00000000..4944db3d
--- /dev/null
+++ b/test/de/lmu/ifi/dbs/elki/evaluation/TestPairCountingFMeasure.java
@@ -0,0 +1,78 @@
+package de.lmu.ifi.dbs.elki.evaluation;
+
+import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertTrue;
+
+import org.junit.Test;
+
+import de.lmu.ifi.dbs.elki.JUnit4Test;
+import de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.ByLabelClustering;
+import de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.TrivialAllInOne;
+import de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.TrivialAllNoise;
+import de.lmu.ifi.dbs.elki.data.Clustering;
+import de.lmu.ifi.dbs.elki.data.model.Model;
+import de.lmu.ifi.dbs.elki.data.type.TypeUtil;
+import de.lmu.ifi.dbs.elki.database.Database;
+import de.lmu.ifi.dbs.elki.database.StaticArrayDatabase;
+import de.lmu.ifi.dbs.elki.database.relation.Relation;
+import de.lmu.ifi.dbs.elki.datasource.FileBasedDatabaseConnection;
+import de.lmu.ifi.dbs.elki.evaluation.paircounting.PairCountingFMeasure;
+import de.lmu.ifi.dbs.elki.utilities.ClassGenericsUtil;
+import de.lmu.ifi.dbs.elki.utilities.optionhandling.ParameterException;
+import de.lmu.ifi.dbs.elki.utilities.optionhandling.parameterization.ListParameterization;
+
+/**
+ * Validate {@link PairCountingFMeasure} with respect to its ability to compare
+ * data clusterings.
+ * 
+ * @author Erich Schubert
+ */
+public class TestPairCountingFMeasure implements JUnit4Test {
+  // the following values depend on the data set used!
+  String dataset = "data/testdata/unittests/hierarchical-3d2d1d.csv";
+
+  // size of the data set
+  int shoulds = 600;
+
+  /**
+   * Validate {@link PairCountingFMeasure} with respect to its ability to
+   * compare data clusterings.
+   * 
+   * @throws ParameterException on errors.
+   */
+  @Test
+  public void testCompareDatabases() {
+    ListParameterization params = new ListParameterization();
+    // Input
+    params.addParameter(FileBasedDatabaseConnection.INPUT_ID, dataset);
+
+    // get database
+    Database db = ClassGenericsUtil.parameterizeOrAbort(StaticArrayDatabase.class, params);
+    db.initialize();
+
+    // verify data set size.
+    Relation<?> rel = db.getRelation(TypeUtil.ANY);
+    assertTrue(rel.size() == shoulds);
+
+    // run all-in-one
+    TrivialAllInOne allinone = new TrivialAllInOne();
+    Clustering<Model> rai = allinone.run(db);
+
+    // run all-in-noise
+    TrivialAllNoise allinnoise = new TrivialAllNoise();
+    Clustering<Model> ran = allinnoise.run(db);
+
+    // run by-label
+    ByLabelClustering bylabel = new ByLabelClustering();
+    Clustering<?> rbl = bylabel.run(db);
+
+    assertEquals(1.0, PairCountingFMeasure.compareClusterings(rai, rai), Double.MIN_VALUE);
+    assertEquals(1.0, PairCountingFMeasure.compareClusterings(ran, ran), Double.MIN_VALUE);
+    assertEquals(1.0, PairCountingFMeasure.compareClusterings(rbl, rbl), Double.MIN_VALUE);
+
+    assertEquals(0.009950248756218905, PairCountingFMeasure.compareClusterings(ran, rbl, true, false), Double.MIN_VALUE);
+    assertEquals(0.0033277870216306157, PairCountingFMeasure.compareClusterings(rai, ran, true, false), Double.MIN_VALUE);
+
+    assertEquals(0.5 /* 0.3834296724470135 */, PairCountingFMeasure.compareClusterings(rai, rbl), Double.MIN_VALUE);
+  }
+}
+\ No newline at end of file