Class PCAFilteredResult


  • public class PCAFilteredResult
    extends PCAResult
    Result class for a filtered PCA. This differs from regular PCA by having the Eigenvalues and Eigenvectors separated into "strong" and "weak" Eigenvectors, and thus a dimension. Usually this will be interpreted as having a "data" subspace and an "error" subspace.
    Since:
    0.2
    Author:
    Erich Schubert
    • Field Summary

      Fields 
      Modifier and Type Field Description
      private double explainedVariance
      The amount of Variance explained by strong Eigenvalues
      private double[][] m_czech
      The dissimilarity matrix.
      private double[][] m_hat
      The similarity matrix.
      private double[] strongEigenvalues
      The strong eigenvalues.
      private double[][] strongEigenvectors
      The strong eigenvectors to their corresponding filtered eigenvalues.
      private double[] weakEigenvalues
      The weak eigenvalues.
      private double[][] weakEigenvectors
      The weak eigenvectors to their corresponding filtered eigenvalues.
    • Constructor Summary

      Constructors 
      Constructor Description
      PCAFilteredResult​(EigenPair[] eigenPairs, int numstrong, double big, double small)
      Construct a result object for the filtered PCA result.
    • Field Detail

      • strongEigenvalues

        private double[] strongEigenvalues
        The strong eigenvalues.
      • strongEigenvectors

        private double[][] strongEigenvectors
        The strong eigenvectors to their corresponding filtered eigenvalues.
      • weakEigenvalues

        private double[] weakEigenvalues
        The weak eigenvalues.
      • weakEigenvectors

        private double[][] weakEigenvectors
        The weak eigenvectors to their corresponding filtered eigenvalues.
      • explainedVariance

        private double explainedVariance
        The amount of Variance explained by strong Eigenvalues
      • m_hat

        private double[][] m_hat
        The similarity matrix.
      • m_czech

        private double[][] m_czech
        The dissimilarity matrix.
    • Constructor Detail

      • PCAFilteredResult

        public PCAFilteredResult​(EigenPair[] eigenPairs,
                                 int numstrong,
                                 double big,
                                 double small)
        Construct a result object for the filtered PCA result.
        Parameters:
        eigenPairs - All EigenPairs
        numstrong - Number of strong eigenvalues
        big - large value in selection matrix
        small - small value in selection matrix
    • Method Detail

      • getStrongEigenvectors

        public final double[][] getStrongEigenvectors()
        Returns the matrix of strong eigenvectors after passing the eigen pair filter.
        Returns:
        the matrix of eigenvectors
      • getStrongEigenvalues

        public final double[] getStrongEigenvalues()
        Returns the strong eigenvalues of the object after passing the eigen pair filter.
        Returns:
        the eigenvalues
      • getWeakEigenvectors

        public final double[][] getWeakEigenvectors()
        Returns the matrix of weak eigenvectors after passing the eigen pair filter.
        Returns:
        the matrix of eigenvectors
      • getWeakEigenvalues

        public final double[] getWeakEigenvalues()
        Returns the weak eigenvalues of the object after passing the eigen pair filter.
        Returns:
        the eigenvalues
      • getCorrelationDimension

        public final int getCorrelationDimension()
        Get correlation (subspace) dimensionality
        Returns:
        length of strong eigenvalues
      • getExplainedVariance

        public double getExplainedVariance()
        Returns explained variance
        Returns:
        the variance explained by the strong Eigenvectors
      • similarityMatrix

        public double[][] similarityMatrix()
        Returns the similarity matrix (M_hat) of this LocalPCA.
        Returns:
        the similarity matrix M_hat
      • dissimilarityMatrix

        public double[][] dissimilarityMatrix()
        Returns the dissimilarity matrix (M_czech) of this LocalPCA.
        Returns:
        the dissimilarity matrix M_hat