CorrelationMIHybrid<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Hybrid

Correlation-Mutual Information Hybrid feature selection.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> โ†’ CorrelationMIHybrid<T>

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Correlation finds straight-line relationships while mutual information finds any type of pattern. A feature might be strongly related to the target in a curved way that correlation misses. This hybrid catches both types of useful features.

How It Works

Combines linear correlation (Pearson) with mutual information (non-linear) to capture both types of feature-target relationships.

Constructors

Constructor Summary
CorrelationMIHybrid(int nFeaturesToSelect, double correlationWeight, int nBins, int[] columnIndices)

Properties

Property Summary
Vector<T> CorrelationScores { get; }
Vector<T> HybridScores { get; }
Vector<T> MIScores { get; }
bool SupportsInverseTransform { get; }
double CorrelationWeight { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

Method Summary
Matrix<T> FitTransform(Matrix<T> data, Vector<T> target)
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)
void Fit(Matrix<T> data, Vector<T> target)