Class ValidateApproximativeKNNIndex<O>

  • Type Parameters:
    O - Object type

    public class ValidateApproximativeKNNIndex<O>
    extends AbstractDistanceBasedApplication<O>
    Algorithm to validate the quality of an approximative kNN index, by performing a number of queries and comparing them to the results obtained by exact indexing (e.g., linear scanning).
    Since:
    0.6.0
    Author:
    Erich Schubert
    • Field Detail

      • LOG

        private static final Logging LOG
        The logger for this class.
      • k

        protected int k
        Number of neighbors to retrieve.
      • queries

        protected DatabaseConnection queries
        The alternate query point source. Optional.
      • sampling

        protected double sampling
        Sampling size.
      • forcelinear

        protected boolean forcelinear
        Force linear scanning.
      • random

        protected RandomFactory random
        Random generator factory
      • pattern

        protected java.util.regex.Pattern pattern
        Filter pattern
    • Constructor Detail

      • ValidateApproximativeKNNIndex

        public ValidateApproximativeKNNIndex​(InputStep input,
                                             Distance<? super O> distance,
                                             int k,
                                             DatabaseConnection queries,
                                             double sampling,
                                             boolean forcelinear,
                                             RandomFactory random,
                                             java.util.regex.Pattern pattern)
        Constructor.
        Parameters:
        input - Data input
        distance - Distance function to use
        k - K parameter
        queries - Query data set (may be null!)
        sampling - Sampling rate
        random - Random factory
        forcelinear - Force the use of linear scanning.
        pattern -