ZScoreDetector<T>

Models & Types · AiDotNet.AnomalyDetection.Statistical

Detects anomalies using the Z-Score method (standard score).

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

Implements: IGeneratedParameterRegistrar<T>

For Beginners

The Z-Score measures how many standard deviations a value is from the mean. A Z-Score of 0 means the value equals the mean. A Z-Score of 2 means the value is 2 standard deviations above the mean. Points with extreme Z-Scores (typically |Z| > 3) are anomalies.

How It Works

When to use:

  • Your data is approximately normally distributed
  • You want a simple, interpretable method
  • You have single-variable or multi-variable data where anomalies are extreme in at least one dimension

Industry Standard Defaults:

  • Threshold: 3.0 (flags ~0.3% of normally distributed data)
  • Contamination: 0.1 (10%) - used for automatic threshold tuning

Constructors

Constructor Summary
ZScoreDetector(double zThreshold, double contamination, int randomSeed) Creates a new Z-Score anomaly detector.

Properties

Property Summary
Vector<T> Means { get; } Gets the fitted mean values for each feature.
Vector<T> StandardDeviations { get; } Gets the fitted standard deviation values for each feature.
double ZThreshold { get; } Gets the Z-Score threshold.

Methods

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