KNNDetector<T>

Models & Types · AiDotNet.AnomalyDetection.DistanceBased

Detects anomalies using K-Nearest Neighbors distance.

Inheritance: ModelBase<T, Matrix<T>, Vector<T>> → AnomalyDetectorBase<T> → KNNDetector<T>

Implements: IGeneratedParameterRegistrar<T>

For Beginners

KNN anomaly detection identifies outliers based on their distance to their k-nearest neighbors. Points that are far from their nearest neighbors are considered anomalies.

How It Works

The algorithm works by:

  1. For each point, find the k nearest neighbors
  2. Calculate the average distance to these neighbors
  3. Points with large average distances are anomalies

When to use:

  • When anomalies are expected to be isolated from normal clusters
  • Works well with low-to-medium dimensional data
  • No assumption about data distribution

Industry Standard Defaults:

  • K (neighbors): 5 (typical range 3-20)
  • Contamination: 0.1 (10%)

Constructors

Constructor Summary
KNNDetector(int k, double contamination, int randomSeed) Creates a new KNN anomaly detector.

Properties

Property Summary
int K { get; } Gets the number of neighbors used for detection.

Methods

Method Summary
Vector<T> ScoreAnomalies(Matrix<T> X)
void Fit(Matrix<T> X)