MultiClusterFS<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter.Spectral
Multi-Cluster Feature Selection (MCFS) using spectral analysis for unsupervised feature selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → MultiClusterFS<T>
Implements: IFixedCountFeatureSelector<T>
For Beginners
MCFS first discovers natural groups (clusters) in your data without knowing the labels. Then it picks features that best explain these groupings. If a feature helps distinguish between clusters, it’s selected. This is useful when you have no labels but want to find discriminative features.
How It Works
MCFS uses spectral clustering to discover the underlying cluster structure of the data, then selects features that best preserve this structure. It combines spectral embedding with sparse regression to select informative features.
Constructors
| Constructor | Summary |
|---|---|
MultiClusterFS(int nFeaturesToSelect, int nClusters, int nNeighbors, double sigma, int? randomState, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> McfsScores { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NClusters { get; } |
|
int NFeaturesToSelect { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |