Examples of ByLabelHierarchicalClustering

de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.ByLabelHierarchicalClustering
Pseudo clustering using labels. This "algorithm" puts elements into the same cluster when they agree in their labels. I.e. it just uses a predefined clustering, and is mostly useful for testing and evaluation (e.g. comparing the result of a real algorithm to a reference result / golden standard). This variant derives a hierarchical result by doing a prefix comparison on labels. TODO: Noise handling (e.g. allow the user to specify a noise label pattern?) @author Erich Schubert @apiviz.uses de.lmu.ifi.dbs.elki.data.ClassLabel

Examples of de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.ByLabelHierarchicalClustering

   * @param clustering Clustering result
   * @param expected Expected score
   */
  protected <O> void testFMeasureHierarchical(Database database, Clustering<?> clustering, double expected) {
    // Run by-label as reference
    ByLabelHierarchicalClustering bylabel = new ByLabelHierarchicalClustering();
    Clustering<Model> rbl = bylabel.run(database);


    double score = PairCountingFMeasure.compareClusterings(clustering, rbl, 1.0, false, true);
    if(logger.isVerbose()) {
      logger.verbose(this.getClass().getSimpleName() + " score: " + score + " expect: " + expected);
    }

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.ByLabelHierarchicalClustering

  private Clustering<Model> generateDefaultClustering() {
    final Database db = ResultUtil.findDatabase(getResult());
    Clustering<Model> c = null;
    try {
      // Try to cluster by labels
      ByLabelHierarchicalClustering split = new ByLabelHierarchicalClustering();
      c = split.run(db);
    }
    catch(NoSupportedDataTypeException e) {
      // Put everything into one
      c = new TrivialAllInOne().run(db);
    }

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.ByLabelHierarchicalClustering

   */
  public static <O> void ensureClusteringResult(final Database db, final Result result) {
    Collection<Clustering<?>> clusterings = ResultUtil.filterResults(result, Clustering.class);
    if(clusterings.size() == 0) {
      try {
        ClusteringAlgorithm<Clustering<Model>> split = new ByLabelHierarchicalClustering();
        Clustering<Model> c = split.run(db);
        addChildResult(db, c);
      }
      catch(NoSupportedDataTypeException e) {
        Clustering<Model> c = (new TrivialAllInOne()).run(db);
        addChildResult(db, c);

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.ByLabelHierarchicalClustering

   */
  public static <O> void ensureClusteringResult(final Database db, final Result result) {
    Collection<Clustering<?>> clusterings = ResultUtil.filterResults(result, Clustering.class);
    if(clusterings.size() == 0) {
      try {
        ClusteringAlgorithm<Clustering<Model>> split = new ByLabelHierarchicalClustering();
        Clustering<Model> c = split.run(db);
        addChildResult(db, c);
      }
      catch(NoSupportedDataTypeException e) {
        Clustering<Model> c = (new TrivialAllInOne()).run(db);
        addChildResult(db, c);

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.algorithm.clustering.trivial.ByLabelHierarchicalClustering

   * 
   * @return generated clustering
   */
  private Clustering<Model> generateDefaultClustering() {
    // Cluster by labels
    ByLabelHierarchicalClustering split = new ByLabelHierarchicalClustering();
    Clustering<Model> c = split.run(ResultUtil.findDatabase(getResult()));
    // store.
    put(CLUSTERING_FALLBACK, c);
    return c;
  }

View Full Code Here

TOP

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.