Class HDBSCANHierarchyExtraction.TempCluster

  • Enclosing class:
    HDBSCANHierarchyExtraction

    private static class HDBSCANHierarchyExtraction.TempCluster
    extends java.lang.Object
    Temporary cluster.
    Author:
    Erich Schubert
    • Field Detail

      • seq

        protected int seq
        Merge id of the cluster for prototype identification.
      • members

        protected ModifiableDBIDs members
        New ids, not yet in child clusters.
      • dist

        protected double dist
        Current height.
      • dmin

        protected double dmin
        Minimum height (densest object).
      • aggregate

        protected double aggregate
        Mass aggregate.
      • childrenTotal

        protected int childrenTotal
        Number of objects in children.
    • Constructor Detail

      • TempCluster

        public TempCluster​(int seq,
                           double dist,
                           DBIDRef a)
        Constructor.
        Parameters:
        seq - Cluster generation sequence
        dist - Distance
        a - Object reference
    • Method Detail

      • grow

        public HDBSCANHierarchyExtraction.TempCluster grow​(int seq,
                                                           double dist,
                                                           DBIDRef id)
        Join the contents of another cluster.
        Parameters:
        seq - Cluster generation sequence
        dist - Join distance
        id - Cluster lead, for 1-element clusters.
        Returns:
        this
      • totalElements

        public int totalElements()
        Total number of elements in this subtree.
        Returns:
        Total
      • excessOfMass

        public double excessOfMass()
        Excess of mass measure.
        Returns:
        Excess of mass
      • totalStability

        public double totalStability()
        Excess of mass measure.
        Returns:
        Excess of mass
      • isSpurious

        public boolean isSpurious​(int minClSize)
        Test if a cluster is spurious.
        Parameters:
        minClSize - Minimum cluster size
        Returns:
        true if spurious