SpectralClusteringFS<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Graph

Spectral Clustering-based Feature Selection.

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

Implements: IFeatureSelector<T>

For Beginners

Spectral clustering looks at features as points in a graph and uses special math (eigenvectors) to find natural groups. We then pick the most representative feature from each group.

How It Works

Uses spectral graph theory to cluster features and select representative features from each cluster based on eigenvector analysis.

Constructors

Constructor Summary
SpectralClusteringFS(int nClusters, double sigma, int[] columnIndices)

Properties

Property Summary
Vector<T> FeatureScores { get; }
bool SupportsInverseTransform { get; }
int NClusters { get; }
int[] ClusterAssignments { get; }
int[] SelectedIndices { get; }

Methods

Method Summary
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)