Package smile.clustering

Class DENCLUE

java.lang.Object
smile.clustering.Partitioning
smile.clustering.DENCLUE
All Implemented Interfaces:
Serializable
public class DENCLUE extends Partitioning
DENsity CLUstering. The DENCLUE algorithm employs a cluster model based on kernel density estimation. A cluster is defined by a local maximum of the estimated density function. Observations going to the same local maximum are put into the same cluster.

Clearly, DENCLUE doesn't work on data with uniform distribution. In high dimensional space, the data always look like uniformly distributed because of the curse of dimensionality. Therefore, DENCLUDE doesn't work well on high-dimensional data in general.

References

  1. A. Hinneburg and D. A. Keim. A general approach to clustering in large databases with noise. Knowledge and Information Systems, 5(4):387-415, 2003.
  2. Alexander Hinneburg and Hans-Henning Gabriel. DENCLUE 2.0: Fast Clustering based on Kernel Density Estimation. IDA, 2007.
See Also:

  • Nested Class Summary

    Nested Classes
    Modifier and Type
    Class
    Description
    static final record 
    DENCLUE.Options
    DENCLUE hyperparameters.

  • Constructor Summary

    Constructors
    Constructor
    Description
    DENCLUE(int k, int[] group, double[][] attractors, double[] radius, double[][] samples, double sigma, double tol)
    Constructor.

  • Method Summary

    Modifier and Type
    Method
    Description
    double[][]
    attractors()
    Returns the density attractor of each data point.
    static DENCLUE
    fit(double[][] data, double sigma, int m)
    Clustering data.
    static DENCLUE
    fit(double[][] data, DENCLUE.Options options)
    Clustering data.
    int
    predict(double[] x)
    Classifies a new observation.
    double[]
    radius()
    Returns the radius of density attractor.
    double
    sigma()
    Returns the smooth parameter in the Gaussian kernel.
    double
    tolerance()
    Returns the tolerance of hill-climbing procedure.

    Methods inherited from class smile.clustering.Partitioning

    group, group, k, size, size, toString

    Methods inherited from class java.lang.Object

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

  • Constructor Details

    • DENCLUE

      public DENCLUE(int k, int[] group, double[][] attractors, double[] radius, double[][] samples, double sigma, double tol)
      Constructor.
      Parameters:
      k - the number of clusters.
      group - the cluster labels.
      attractors - the density attractor of each data point.
      radius - the radius of density attractor.
      samples - the samples in the iterations of hill climbing.
      sigma - the smooth parameter in the Gaussian kernel. The user can choose sigma such that number of density attractors is constant for a long interval of sigma.
      tol - the tolerance of hill-climbing procedure.

  • Method Details

    • sigma

      public double sigma()
      Returns the smooth parameter in the Gaussian kernel.
      Returns:
      the smooth parameter in the Gaussian kernel.

    • tolerance

      public double tolerance()
      Returns the tolerance of hill-climbing procedure.
      Returns:
      the tolerance of hill-climbing procedure.

    • radius

      public double[] radius()
      Returns the radius of density attractor.
      Returns:
      the radius of density attractor.

    • attractors

      public double[][] attractors()
      Returns the density attractor of each data point.
      Returns:
      the density attractor of each data point.

    • fit

      public static DENCLUE fit(double[][] data, double sigma, int m)
      Clustering data.
      Parameters:
      data - the input data of which each row is an observation.
      sigma - the smooth parameter in the Gaussian kernel. The user can choose sigma such that number of density attractors is constant for a long interval of sigma.
      m - the number of selected samples used in the iteration. This number should be much smaller than the number of observations to speed up the algorithm. It should also be large enough to capture the sufficient information of underlying distribution.
      Returns:
      the model.

    • fit

      public static DENCLUE fit(double[][] data, DENCLUE.Options options)
      Clustering data.
      Parameters:
      data - the input data of which each row is an observation.
      options - the hyperparameters.
      Returns:
      the model.

    • predict

      public int predict(double[] x)
      Classifies a new observation.
      Parameters:
      x - a new observation.
      Returns:
      the cluster label. Note that it may be Clustering.OUTLIER.