Package de.jungblut.clustering

Class DBSCAN

  • public final class DBSCAN
extends java.lang.Object
    Sequential version of DBSCAN to evaluate if this algorithm is suitable for arbitrary parallelization paradigms that can crunch graphs.

    PLAN:

    1. compute distance matrix between the points
    2. extract adjacent points via threshold epsilon and minpoints s
    3. run connected components (here BFS)
    4. PROFIT!

    • Constructor Summary

      Constructors 
      Constructor Description
      DBSCAN()  

    • Method Summary

      All Methods Instance Methods Concrete Methods 
      Modifier and Type Method Description
      java.util.ArrayList<de.jungblut.math.DoubleVector>[] cluster​(java.util.List<de.jungblut.math.DoubleVector> points, DistanceMeasurer measurer, int minPoints, double epsilon)
      Clusters the points.
      java.util.List<de.jungblut.math.DoubleVector> getNoise()  

      • Methods inherited from class java.lang.Object

        clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

    • Constructor Detail

      • DBSCAN

        public DBSCAN()

    • Method Detail

      • cluster

        public java.util.ArrayList<de.jungblut.math.DoubleVector>[] cluster​(java.util.List<de.jungblut.math.DoubleVector> points,
                                                                    DistanceMeasurer measurer,
                                                                    int minPoints,
                                                                    double epsilon)
        Clusters the points.
        Parameters:
        measurer - the distance measurer to use.
        minPoints - the minimum points in a cluster.
        epsilon - the radius of a point to detect other points.

      • getNoise

        public java.util.List<de.jungblut.math.DoubleVector> getNoise()
        Returns:
        the found noise as list of vectors.