Package elki.clustering.hierarchical

Class SLINK<O>

  • Type Parameters:
    O - the type of DatabaseObject the algorithm is applied on
    All Implemented Interfaces:
    elki.Algorithm, HierarchicalClusteringAlgorithm
    Direct Known Subclasses:
    CLINK
    @Title("SLINK: Single Link Clustering")
@Description("Hierarchical clustering algorithm based on single-link connectivity.")
@Reference(authors="R. Sibson",
           title="SLINK: An optimally efficient algorithm for the single-link cluster method",
           booktitle="The Computer Journal 16 (1)",
           url="https://doi.org/10.1093/comjnl/16.1.30",
           bibkey="DBLP:journals/cj/Sibson73")
@Alias({"single-link","single-linkage"})
@Priority(200)
public class SLINK<O>
extends java.lang.Object
implements HierarchicalClusteringAlgorithm
    Implementation of the efficient Single-Link Algorithm SLINK of R. Sibson.

    For ELKI 0.8.0, this was rewritten to no longer use the original "pointer" output format, but instead generate an easier to use merge history now.

    This is probably the fastest exact single-link algorithm currently in use.

    Reference:

    R. Sibson
    SLINK: An optimally efficient algorithm for the single-link cluster method
    In: The Computer Journal 16 (1)

    Since:
    0.6.0
    Author:
    Elke Achtert, Erich Schubert

    • Nested Class Summary

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

      • Nested classes/interfaces inherited from interface elki.Algorithm

        elki.Algorithm.Utils

    • Field Summary

      Fields 
      Modifier and Type Field Description
      protected elki.distance.Distance<? super O> distance
      Distance function used.
      private static elki.logging.Logging LOG
      The logger for this class.

    • Constructor Summary

      Constructors 
      Constructor Description
      SLINK​(elki.distance.Distance<? super O> distance)
      Constructor.

    • Method Summary

      All Methods Static Methods Instance Methods Concrete Methods 
      Modifier and Type Method Description
      protected static ClusterMergeHistoryBuilder convertOutput​(ClusterMergeHistoryBuilder builder, elki.database.ids.ArrayDBIDs oids, elki.database.datastore.DBIDDataStore pi, elki.database.datastore.DoubleDataStore lambda)
      Convert a SLINK pointer representation to a cluster merge history.
      elki.data.type.TypeInformation[] getInputTypeRestriction()  
      protected elki.logging.Logging getLogger()
      Get the (static) class logger.
      protected void process​(elki.database.ids.DBIDRef id, elki.database.ids.ArrayDBIDs ids, elki.database.ids.DBIDArrayIter it, int n, elki.database.datastore.WritableDBIDDataStore pi, elki.database.datastore.WritableDoubleDataStore lambda, elki.database.datastore.WritableDoubleDataStore m)
      SLINK main loop.
      ClusterMergeHistory run​(elki.database.relation.Relation<O> relation)
      Performs the SLINK algorithm on the given database.
      private void slinkstep3​(elki.database.ids.DBIDRef id, elki.database.ids.DBIDArrayIter it, int n, elki.database.datastore.WritableDBIDDataStore pi, elki.database.datastore.WritableDoubleDataStore lambda, elki.database.datastore.WritableDoubleDataStore m)
      Third step: Determine the values for P and L
      private void slinkstep4​(elki.database.ids.DBIDRef id, elki.database.ids.DBIDArrayIter it, int n, elki.database.datastore.WritableDBIDDataStore pi, elki.database.datastore.WritableDoubleDataStore lambda)
      Fourth step: Actualize the clusters if necessary
      private void step2​(elki.database.ids.DBIDRef id, elki.database.ids.DBIDArrayIter it, int n, elki.database.query.distance.DistanceQuery<? super O> distQuery, elki.database.datastore.WritableDoubleDataStore m)
      Second step: Determine the pairwise distances from all objects in the pointer representation to the new object with the specified id.
      private void step2primitive​(elki.database.ids.DBIDRef id, elki.database.ids.DBIDArrayIter it, int n, elki.database.relation.Relation<? extends O> relation, elki.distance.PrimitiveDistance<? super O> distance, elki.database.datastore.WritableDoubleDataStore m)
      Second step: Determine the pairwise distances from all objects in the pointer representation to the new object with the specified id.

      • Methods inherited from class java.lang.Object

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

    • Field Detail

      • LOG

        private static final elki.logging.Logging LOG
        The logger for this class.

      • distance

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

    • Constructor Detail

      • SLINK

        public SLINK​(elki.distance.Distance<? super O> distance)
        Constructor.
        Parameters:
        distance - Distance function

    • Method Detail

      • getInputTypeRestriction

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

      • run

        public ClusterMergeHistory run​(elki.database.relation.Relation<O> relation)
        Performs the SLINK algorithm on the given database.
        Parameters:
        relation - Data relation to use

      • convertOutput

        protected static ClusterMergeHistoryBuilder convertOutput​(ClusterMergeHistoryBuilder builder,
                                                          elki.database.ids.ArrayDBIDs oids,
                                                          elki.database.datastore.DBIDDataStore pi,
                                                          elki.database.datastore.DoubleDataStore lambda)
        Convert a SLINK pointer representation to a cluster merge history.
        Parameters:
        builder - Builder
        oids - original DBIDs
        pi - Parent pointer
        lambda - Parent distance
        Returns:
        Builder

      • step2

        private void step2​(elki.database.ids.DBIDRef id,
                   elki.database.ids.DBIDArrayIter it,
                   int n,
                   elki.database.query.distance.DistanceQuery<? super O> distQuery,
                   elki.database.datastore.WritableDoubleDataStore m)
        Second step: Determine the pairwise distances from all objects in the pointer representation to the new object with the specified id.
        Parameters:
        id - the id of the object to be inserted into the pointer representation
        it - Array iterator
        n - Last object
        distQuery - Distance query
        m - Data store

      • step2primitive

        private void step2primitive​(elki.database.ids.DBIDRef id,
                            elki.database.ids.DBIDArrayIter it,
                            int n,
                            elki.database.relation.Relation<? extends O> relation,
                            elki.distance.PrimitiveDistance<? super O> distance,
                            elki.database.datastore.WritableDoubleDataStore m)
        Second step: Determine the pairwise distances from all objects in the pointer representation to the new object with the specified id.
        Parameters:
        id - the id of the object to be inserted into the pointer representation
        it - Array iterator
        n - Last object
        m - Data store
        relation - Data relation
        distance - Distance function to use

      • process

        protected void process​(elki.database.ids.DBIDRef id,
                       elki.database.ids.ArrayDBIDs ids,
                       elki.database.ids.DBIDArrayIter it,
                       int n,
                       elki.database.datastore.WritableDBIDDataStore pi,
                       elki.database.datastore.WritableDoubleDataStore lambda,
                       elki.database.datastore.WritableDoubleDataStore m)
        SLINK main loop.
        Parameters:
        id - Current object
        ids - All objects
        it - Array iterator
        n - Last object to process at this run
        pi - Parent
        lambda - Height
        m - Distance

      • slinkstep3

        private void slinkstep3​(elki.database.ids.DBIDRef id,
                        elki.database.ids.DBIDArrayIter it,
                        int n,
                        elki.database.datastore.WritableDBIDDataStore pi,
                        elki.database.datastore.WritableDoubleDataStore lambda,
                        elki.database.datastore.WritableDoubleDataStore m)
        Third step: Determine the values for P and L
        Parameters:
        id - the id of the object to be inserted into the pointer representation
        it - array iterator
        n - Last object to process at this run
        pi - Pi data store
        lambda - Lambda data store
        m - Data store

      • slinkstep4

        private void slinkstep4​(elki.database.ids.DBIDRef id,
                        elki.database.ids.DBIDArrayIter it,
                        int n,
                        elki.database.datastore.WritableDBIDDataStore pi,
                        elki.database.datastore.WritableDoubleDataStore lambda)
        Fourth step: Actualize the clusters if necessary
        Parameters:
        id - the id of the current object
        it - array iterator
        n - Last object to process at this run
        pi - Pi data store
        lambda - Lambda data store

      • getLogger

        protected elki.logging.Logging getLogger()
        Get the (static) class logger.