Class DBSCANOutlierDetection

  • All Implemented Interfaces:
    Algorithm, OutlierAlgorithm

    @Title("DBSCAN Outlier Detection: Outlier Detection based on the Generalized DBSCAN clustering")
    public class DBSCANOutlierDetection
    extends java.lang.Object
    implements OutlierAlgorithm
    Outlier detection algorithm using DBSCAN Clustering.

    The outlierness score is attributed to each point according to the set to which it is linked: global outliers (noise points), local outliers (border points), and inliers (core points).

    Since:
    0.7.5
    Author:
    Braulio V.S. Vinces
    • Constructor Detail

      • DBSCANOutlierDetection

        public DBSCANOutlierDetection​(GeneralizedDBSCAN clusterer)
        Constructor with an existing Genearalized DBSCAN clustering algorithm.
        Parameters:
        clusterer - Generalized DBSCAN clustering algorithm to use.
    • Method Detail

      • run

        public OutlierResult run​(Database db,
                                 Relation<? extends NumberVector> relation)
        Runs the algorithm in the timed evaluation part.
        Parameters:
        relation - Relation to process
        Returns:
        Outlier result
      • getInputTypeRestriction

        public TypeInformation[] getInputTypeRestriction()
        Description copied from interface: Algorithm
        Get the input type restriction used for negotiating the data query.
        Specified by:
        getInputTypeRestriction in interface Algorithm
        Returns:
        Type restriction