MutualInformationClassification<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Filter.Univariate

Mutual Information for classification feature selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Mutual information measures how much knowing a feature tells you about the class. If knowing the feature value significantly reduces uncertainty about the class, the mutual information is high. Unlike correlation, it can capture any type of relationship, not just linear ones.

How It Works

Measures the mutual information between each feature and the target class. Captures any kind of dependency (linear or nonlinear) between features and target.

Constructors

Constructor Summary
MutualInformationClassification(int nFeaturesToSelect, int nBins, bool discreteFeatures, int[] columnIndices)

Properties

Property Summary
Vector<T> MutualInfoScores { get; }
bool SupportsInverseTransform { 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)