Class PCA

java.lang.Object
smile.feature.extraction.Projection
smile.feature.extraction.PCA
All Implemented Interfaces:
Serializable, Function<Tuple,Tuple>, Transform
public class PCA extends Projection
Principal component analysis. PCA is an orthogonal linear transformation that transforms a number of possibly correlated variables into a smaller number of uncorrelated variables called principal components. The first principal component accounts for as much of the variability in the data as possible, and each succeeding component accounts for as much of the remaining variability as possible. PCA is theoretically the optimum transform for given data in least square terms. PCA can be thought of as revealing the internal structure of the data in a way which best explains the variance in the data. If a multivariate dataset is visualized as a set of coordinates in a high-dimensional data space, PCA supplies the user with a lower-dimensional picture when viewed from its (in some sense) most informative viewpoint.

PCA is mostly used as a tool in exploratory data analysis and for making predictive models. PCA involves the calculation of the eigenvalue decomposition of a data covariance matrix or singular value decomposition of a data matrix, usually after mean centering the data for each attribute. The results of a PCA are usually discussed in terms of component scores and loadings.

As a linear technique, PCA is built for several purposes: first, it enables us to decorrelate the original variables; second, to carry out data compression, where we pay decreasing attention to the numerical accuracy by which we encode the sequence of principal components; third, to reconstruct the original input data using a reduced number of variables according to a least-squares criterion; and fourth, to identify potential clusters in the data.

In certain applications, PCA can be misleading. PCA is heavily influenced when there are outliers in the data. In other situations, the linearity of PCA may be an obstacle to successful data reduction and compression.

See Also:

  • Constructor Details

    • PCA

      public PCA(Vector mu, Vector eigvalues, DenseMatrix loadings, DenseMatrix projection, String... columns)
      Constructor.
      Parameters:
      mu - the mean of samples.
      eigvalues - the eigen values of principal components.
      loadings - the matrix of variable loadings.
      projection - the projection matrix.
      columns - the columns to transform when applied on Tuple/DataFrame.

  • Method Details

    • fit

      public static PCA fit(DataFrame data, String... columns)
      Fits principal component analysis with covariance matrix.
      Parameters:
      data - training data of which each row is a sample.
      columns - the columns to fit PCA. If empty, all columns will be used.
      Returns:
      the model.

    • cor

      public static PCA cor(DataFrame data, String... columns)
      Fits principal component analysis with correlation matrix.
      Parameters:
      data - training data of which each row is a sample.
      columns - the columns to fit PCA. If empty, all columns will be used.
      Returns:
      the model.

    • fit

      public static PCA fit(double[][] data, String... columns)
      Fits principal component analysis with covariance matrix.
      Parameters:
      data - training data of which each row is a sample.
      columns - the columns to transform when applied on Tuple/DataFrame.
      Returns:
      the model.

    • cor

      public static PCA cor(double[][] data, String... columns)
      Fits principal component analysis with correlation matrix.
      Parameters:
      data - training data of which each row is a sample.
      columns - the columns to transform when applied on Tuple/DataFrame.
      Returns:
      the model.

    • center

      public Vector center()
      Returns the center of data.
      Returns:
      the center of data.

    • loadings

      public DenseMatrix loadings()
      Returns the variable loading matrix, ordered from largest to smallest by corresponding eigenvalues. The matrix columns contain the eigenvectors.
      Returns:
      the variable loading matrix.

    • variance

      public Vector variance()
      Returns the principal component variances, ordered from largest to smallest, which are the eigenvalues of the covariance or correlation matrix of learning data.
      Returns:
      the principal component variances.

    • varianceProportion

      public Vector varianceProportion()
      Returns the proportion of variance contained in each principal component, ordered from largest to smallest.
      Returns:
      the proportion of variance contained in each principal component.

    • cumulativeVarianceProportion

      public Vector cumulativeVarianceProportion()
      Returns the cumulative proportion of variance contained in principal components, ordered from largest to smallest.
      Returns:
      the cumulative proportion of variance.

    • getProjection

      public PCA getProjection(int p)
      Returns the projection with given number of principal components.
      Parameters:
      p - choose top p principal components used for projection.
      Returns:
      a new PCA projection.

    • getProjection

      public PCA getProjection(double p)
      Returns the projection with top principal components that contain (more than) the given percentage of variance.
      Parameters:
      p - the required percentage of variance.
      Returns:
      a new PCA projection.

    • postprocess

      protected double[] postprocess(double[] x)
      Description copied from class: Projection
      Postprocess the output vector after projection.
      Overrides:
      postprocess in class Projection
      Parameters:
      x - the output vector of projection.
      Returns:
      the postprocessed vector.