PrincipalComponentAnalysis Class

public sealed class PrincipalComponentAnalysis : TransformationModel<double>

Public NotInheritable Class PrincipalComponentAnalysis
	Inherits TransformationModel(Of Double)

public ref class PrincipalComponentAnalysis sealed : public TransformationModel<double>

[<SealedAttribute>]
type PrincipalComponentAnalysis = 
    class
        inherit TransformationModel<float>
    end

Inheritance

Object  →  Model  →  TransformationModel<Double>  →  PrincipalComponentAnalysis

Use the PrincipalComponentAnalysis class to obtain information about the components that contribute the most to the variation in a dataset. The data for the analysis can be supplied in a variety of formats, including as a data frame, an array of numerical vectors, or a matrix.

The Standardize property determines whether each variable should be rescaled to have zero mean and unit standard deviation. The Fit() method performs the actual calculation.

Once the analysis has been completed, the Components property provides access to a collection of PrincipalComponent objects that provide details about each of the principal components. The components are sorted in order of their contribution to the variance in the data, in descending order. The ComponentMatrix property returns the components as the columns of a matrix. The ScoreMatrix property expresses the observations in terms of the components. The GetPredictions(Int32) method returns the observations if only the specified number of components is taken into account.

The VarianceProportions and CumulativeVarianceProportions summarize the contribution of each component. The GetVarianceThreshold(Double) calculates how many components are needed to explain a certain proportion of the variation.

Reference