Package elki.outlier.lof

Class LOCI<O>

  • Type Parameters:
    O - Object type
    All Implemented Interfaces:
    elki.Algorithm, OutlierAlgorithm
    @Title("LOCI: Fast Outlier Detection Using the Local Correlation Integral")
@Description("Algorithm to compute outliers based on the Local Correlation Integral")
@Reference(authors="S. Papadimitriou, H. Kitagawa, P. B. Gibbons, C. Faloutsos",
           title="LOCI: Fast Outlier Detection Using the Local Correlation Integral",
           booktitle="Proc. 19th IEEE Int. Conf. on Data Engineering (ICDE \'03)",
           url="https://doi.org/10.1109/ICDE.2003.1260802",
           bibkey="DBLP:conf/icde/PapadimitriouKGF03")
public class LOCI<O>
extends java.lang.Object
implements OutlierAlgorithm
    Fast Outlier Detection Using the "Local Correlation Integral".

    Exact implementation only, not aLOCI. See ALOCI.

    Outlier detection using multiple epsilon neighborhoods.

    This implementation has O(n3 log n) runtime complexity!

    Reference:

    S. Papadimitriou, H. Kitagawa, P. B. Gibbons, C. Faloutsos:
    LOCI: Fast Outlier Detection Using the Local Correlation Integral.
    In: Proc. 19th IEEE Int. Conf. on Data Engineering (ICDE '03)

    Since:
    0.2
    Author:
    Erich Schubert

    • Nested Class Summary

      Nested Classes 
      Modifier and Type Class Description
      private static class  LOCI.DoubleIntArrayList
      Array of double-int values.

      • Nested classes/interfaces inherited from interface elki.Algorithm

        elki.Algorithm.Utils

    • Field Summary

      Fields 
      Modifier and Type Field Description
      private double alpha
      Scaling of averaging neighborhood.
      private elki.distance.Distance<? super O> distance
      Distance function used.
      private static elki.logging.Logging LOG
      The logger for this class.
      private int nmin
      Minimum neighborhood size.
      private double rmax
      Maximum radius.

    • Constructor Summary

      Constructors 
      Constructor Description
      LOCI​(elki.distance.Distance<? super O> distance, double rmax, int nmin, double alpha)
      Constructor.

    • Method Summary

      All Methods Instance Methods Concrete Methods 
      Modifier and Type Method Description
      elki.data.type.TypeInformation[] getInputTypeRestriction()  
      protected void precomputeInterestingRadii​(elki.database.ids.DBIDs ids, elki.database.query.range.RangeSearcher<elki.database.ids.DBIDRef> rangeQuery, elki.database.datastore.WritableDataStore<LOCI.DoubleIntArrayList> interestingDistances)
      Preprocessing step: determine the radii of interest for each point.
      OutlierResult run​(elki.database.relation.Relation<O> relation)
      Run the algorithm

      • 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

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

      • rmax

        private double rmax
        Maximum radius.

      • nmin

        private int nmin
        Minimum neighborhood size.

      • alpha

        private double alpha
        Scaling of averaging neighborhood.

    • Constructor Detail

      • LOCI

        public LOCI​(elki.distance.Distance<? super O> distance,
            double rmax,
            int nmin,
            double alpha)
        Constructor.
        Parameters:
        distance - Distance function
        rmax - Maximum radius
        nmin - Minimum neighborhood size
        alpha - Alpha value

    • Method Detail

      • getInputTypeRestriction

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

      • run

        public OutlierResult run​(elki.database.relation.Relation<O> relation)
        Run the algorithm
        Parameters:
        relation - Relation to process
        Returns:
        Outlier result

      • precomputeInterestingRadii

        protected void precomputeInterestingRadii​(elki.database.ids.DBIDs ids,
                                          elki.database.query.range.RangeSearcher<elki.database.ids.DBIDRef> rangeQuery,
                                          elki.database.datastore.WritableDataStore<LOCI.DoubleIntArrayList> interestingDistances)
        Preprocessing step: determine the radii of interest for each point.
        Parameters:
        ids - IDs to process
        rangeQuery - Range query
        interestingDistances - Distances of interest