Class DisparateImpactBiasDetector<T>

Namespace

AiDotNet.Interpretability

Assembly

AiDotNet.dll

Detects bias using the Disparate Impact metric (80% rule). Disparate Impact Ratio = (Min Positive Rate) / (Max Positive Rate). A ratio below 0.8 indicates potential bias.

public class DisparateImpactBiasDetector<T> : BiasDetectorBase<T>, IBiasDetector<T>

Type Parameters

T

The numeric type for calculations.

Inheritance

object

BiasDetectorBase<T>

DisparateImpactBiasDetector<T>

Implements

IBiasDetector<T>

Inherited Members

BiasDetectorBase<T>._numOps

BiasDetectorBase<T>._isLowerBiasBetter

BiasDetectorBase<T>.DetectBias(Vector<T>, Vector<T>, Vector<T>)

BiasDetectorBase<T>.IsLowerBiasBetter

BiasDetectorBase<T>.IsBetterBiasScore(T, T)

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Constructors

DisparateImpactBiasDetector(double)

Initializes a new instance of the DisparateImpactBiasDetector class.

public DisparateImpactBiasDetector(double threshold = 0.8)

Parameters

threshold double

The threshold for detecting bias (default is 0.8, representing the 80% rule from fair lending guidelines).

Methods

GetBiasDetectionResult(Vector<T>, Vector<T>, Vector<T>?)

Implements bias detection using Disparate Impact ratio (80% rule).

protected override BiasDetectionResult<T> GetBiasDetectionResult(Vector<T> predictions, Vector<T> sensitiveFeature, Vector<T>? actualLabels)

Parameters

predictions Vector<T>

sensitiveFeature Vector<T>

actualLabels Vector<T>

Returns

BiasDetectionResult<T>