Package elki.clustering.kcenter

Class GreedyKCenter<O>

  • Type Parameters:
    O - Object type to cluster
    All Implemented Interfaces:
    elki.Algorithm, ClusteringAlgorithm<Clustering<SimplePrototypeModel<O>>>
    @Title("Greedy K-center Clustering")
@Reference(authors="T. F. Gonzalez",title="Clustering to Minimize the Maximum Intercluster Distance",booktitle="Theoretical Computer Science, 38",url="https://doi.org/10.1016/0304-3975(85)90224-5",bibkey="DBLP:journals/tcs/Gonzalez85") @Reference(authors="D. S. Hochbaum, D. B. Shmoys",title="A unified approach to approximation algorithms for bottleneck problems",booktitle="Journal of the ACM, 33 (3), 1986",url="https://doi.org/10.1145/5925.5933",bibkey="DBLP:journals/jacm/HochbaumS86")
public class GreedyKCenter<O>
extends java.lang.Object
implements ClusteringAlgorithm<Clustering<SimplePrototypeModel<O>>>
    Greedy algorithm for k-center algorithm also known as Gonzalez clustering, or farthest-first traversal.

    The first cluster center is chosen arbitrarily. The remaining are always chosen such as to maximize the distance to their current cluster.

    Reference:

    T. F. Gonzalez:
    Clustering to Minimize the Maximum Intercluster Distance
    Theoretical Computer Science, 38, 1985

    D. S. Hochbaum, D. B. Shmoys
    A unified approach to approximation algorithms for bottleneck problems
    Journal of the ACM, 33 (3), 1986

    Since:
    0.8.0
    Author:
    Robert Gehde, Erich Schubert

    • Nested Class Summary

      Nested Classes 
      Modifier and Type Class Description
      static class  GreedyKCenter.Par<O>
      Parameterization class

      • Nested classes/interfaces inherited from interface elki.Algorithm

        elki.Algorithm.Utils

    • Field Summary

      Fields 
      Modifier and Type Field Description
      (package private) elki.distance.Distance<? super O> distance
      Distance function
      (package private) int k
      number of clusters
      private static elki.logging.Logging LOG
      Class logger
      (package private) elki.utilities.random.RandomFactory rand
      Random factory for choosing the first element.

    • Constructor Summary

      Constructors 
      Constructor Description
      GreedyKCenter​(int k, elki.distance.Distance<? super O> distance, elki.utilities.random.RandomFactory rand)
      Constructor.

    • Field Detail

      • LOG

        private static final elki.logging.Logging LOG
        Class logger

      • distance

        elki.distance.Distance<? super O> distance
        Distance function

      • k

        int k
        number of clusters

      • rand

        elki.utilities.random.RandomFactory rand
        Random factory for choosing the first element.

    • Constructor Detail

      • GreedyKCenter

        public GreedyKCenter​(int k,
                     elki.distance.Distance<? super O> distance,
                     elki.utilities.random.RandomFactory rand)
        Constructor.
        Parameters:
        k - number of clusters
        distance - distance function to use
        rand - random factory

    • Method Detail

      • run

        public Clustering<SimplePrototypeModel<O>> run​(elki.database.relation.Relation<O> relation)
        Perform greedy k-center clustering on the relation.
        Parameters:
        relation - data to cluster
        Returns:
        clustering

      • getInputTypeRestriction

        public elki.data.type.TypeInformation[] getInputTypeRestriction()
        Specified by:
        getInputTypeRestriction in interface elki.Algorithm