Class LiftChart

Represents a lift chart, used to measure predictive performance.

Inheritance

System.Object

LiftChart

Inherited Members

System.Object.Equals(System.Object)

System.Object.Equals(System.Object, System.Object)

System.Object.GetHashCode()

System.Object.GetType()

System.Object.MemberwiseClone()

System.Object.ReferenceEquals(System.Object, System.Object)

System.Object.ToString()

Namespace: BayesServer.Analysis

Assembly: BayesServer.Analysis.dll

Syntax

[Obfuscation(Exclude = false)]
public sealed class LiftChart

Remarks

Lift charts are used to measure the predictive performance of a classification network, and tell us how the model compares to a random model and the ideal model.

Lift charts are particularly useful when the probabilities of correct classification are low, such as when predicting rare events or anomalies. In these circumstances, a confusion matrix may give very high accuracy simply because most cases are not rare events or anomalies and hence correctly predicted. A lift chart however can take small probabilities into account to assess the improvement (lift) in prediction of these rare events or anomalies.

Properties

Actual

Gets the name of the data column containing the actual classification.

Declaration

public string Actual { get; }

Property Value

Type	Description
System.String

Ideal

Gets the population probability value at which the target reaches 100 %.

Declaration

public double Ideal { get; }

Property Value

Type	Description
System.Double

Points

Gets the xy points that make up the lift chart.

Declaration

public IList<LiftChartPoint> Points { get; }

Property Value

Type	Description
System.Collections.Generic.IList<LiftChartPoint>

PredictedProbability

Gets the name of the data column which contains the predicted probability generated by an inference query.

Declaration

public string PredictedProbability { get; }

Property Value

Type	Description
System.String

Score

Gets the overall score, which is a positive or negative probability between 0 and 1 indicating the classification performance of the network.

Declaration

public double Score { get; }

Property Value

Type	Description
System.Double

Remarks

Score = (area under predicted - area under random) / (area under ideal - area under random)

A score of 0 indicates that the predictive performance is no better than random.

A score of 1 indicates the network perfectly predicts all cases.

A negative score indicates the network performs worse than random, which may indicate incorrect usage.

TargetValue

Gets the target value which we are interested in.

Declaration

public IComparable TargetValue { get; }

Property Value

Type	Description
System.IComparable

Remarks

For example, for a variable 'Purchase' with state values True and False, if we are trying to predict whether someone will make a purchase or not, the target value would be True.

Methods

Create(String, String, String, IComparable, IDataReaderCommand)

Creates a lift chart, used to measure predictive performance.

Declaration

public static LiftChart Create(string actual, string predictedProbability, string caseWeight, IComparable targetValue, IDataReaderCommand readerCommand)

Parameters

Type	Name	Description
System.String	actual	The name of the data column containing the actual classification.
System.String	predictedProbability	The name of the data column which contains the predicted probability generated by an inference query.
System.String	caseWeight	Optional case weight. Can be null.
System.IComparable	targetValue	The target value of interest. E.g. True if we are interested in predicting Purchase=True.
IDataReaderCommand	readerCommand	The predictions.

Returns

Type	Description
LiftChart	A lift chart.