LOFBasedSelector<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Density
Local Outlier Factor (LOF) based Feature Selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → LOFBasedSelector<T>
Implements: IFixedCountFeatureSelector<T>
For Beginners
LOF measures how “outlying” each point is compared to its neighbors. This selector chooses features that help distinguish normal points from outliers, preserving the local density patterns in your data.
How It Works
Uses the Local Outlier Factor concept to select features that best preserve local density structure, helping maintain meaningful neighborhoods.
Constructors
| Constructor | Summary |
|---|---|
LOFBasedSelector(int nFeaturesToSelect, int nNeighbors, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> LOFContributions { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NFeaturesToSelect { get; } |
|
int NNeighbors { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |