JointMutualInformation<T>

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

Joint Mutual Information (JMI) for feature selection.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Instead of just picking features that are individually informative, JMI picks features that work well together. Two features might each be only moderately useful alone, but together they might be very powerful.

How It Works

JMI maximizes the joint mutual information between selected features and the target. It considers both the relevance of individual features and their complementary information when combined with already selected features.

Constructors

Constructor Summary
JointMutualInformation(int nFeaturesToSelect, int nBins, int[] columnIndices)

Properties

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