GraphBasedFS<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Graph

Graph-based Feature Selection using feature relationships.

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

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Imagine features as people in a social network. Features that are “well connected” (correlated with many other features) might be key features. We use social network analysis techniques to find the most “influential” features.

How It Works

Constructs a graph where nodes are features and edges represent relationships (like correlation). Uses graph centrality measures to identify important features.

Constructors

Constructor Summary
GraphBasedFS(int nFeaturesToSelect, double edgeThreshold, GraphCentralityMeasure centralityMeasure, int[] columnIndices)

Properties

Property Summary
Vector<T> CentralityScores { 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)