FCBF<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter.InformationTheory

Fast Correlation-Based Filter (FCBF) for feature selection.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → FCBF<T>

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Having two features that say the same thing is wasteful. FCBF first picks features that are useful for prediction, then removes features that duplicate what other selected features already tell you. It’s like building a team where each member brings unique skills.

How It Works

FCBF uses Symmetric Uncertainty to evaluate feature-target and feature-feature correlations. It selects features that are highly correlated with the target but not redundant with other selected features, using a fast backwards elimination process.

Constructors

Constructor Summary
FCBF(int nFeaturesToSelect, double threshold, int nBins, int[] columnIndices)

Properties

Property Summary
Vector<T> SUScores { get; }
bool SupportsInverseTransform { get; }
double Threshold { get; }
int NBins { 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)