Package edu.cmu.tetrad.search.unmix

Class UnmixDiagnostics.StabilityResult

java.lang.Object

edu.cmu.tetrad.search.unmix.UnmixDiagnostics.StabilityResult

Enclosing class:

UnmixDiagnostics

public static final class UnmixDiagnostics.StabilityResult
extends Object

Represents the result of a stability analysis of clustering, typically performed by evaluating the Adjusted Rand Index (ARI) across multiple independent runs.

The stability analysis provides:

• The mean Adjusted Rand Index (meanARI) across all pairwise comparisons of cluster labelings.
• The standard deviation of the Adjusted Rand Index (sdARI) across these comparisons.
• The total number of pairwise comparisons (numPairs) considered.

This data structure is used to summarize clustering stability, often in the context of algorithms like Expectation-Maximization or similar clustering methods.

Field Summary

Fields

Modifier and Type	Field	Description
final double	meanARI	Represents the mean Adjusted Rand Index (ARI) across all pairwise comparisons of cluster labelings in a clustering stability analysis.
final int	numPairs	Represents the total number of pairwise comparisons considered in the context of a clustering stability analysis.
final double	sdARI	Represents the standard deviation of the Adjusted Rand Index (ARI) across multiple pairwise comparisons of cluster labelings in a clustering stability analysis.

Constructor Summary

Constructors

Constructor	Description
StabilityResult(double meanARI, double sdARI, int numPairs)	Constructs an object to represent the result of a stability analysis of clustering.

Method Summary

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

meanARI

public final double meanARI

Represents the mean Adjusted Rand Index (ARI) across all pairwise comparisons of cluster labelings in a clustering stability analysis.

The ARI is a measure of similarity between two cluster labelings, adjusted for chance. `meanARI` provides an overall metric summarizing the stability of a clustering algorithm by averaging the ARI values across multiple independent runs.

sdARI

public final double sdARI

Represents the standard deviation of the Adjusted Rand Index (ARI) across multiple pairwise comparisons of cluster labelings in a clustering stability analysis.

This metric quantifies the variability of ARI values, providing insight into the consistency of clustering results across independent runs. A lower value of `sdARI` indicates more stability, while a higher value suggests greater variability in clustering outcomes.

numPairs

public final int numPairs

Represents the total number of pairwise comparisons considered in the context of a clustering stability analysis. This value is used to quantify the number of evaluations performed when computing metrics such as the Adjusted Rand Index (ARI) across multiple independent runs of a clustering algorithm.

Constructor Details

StabilityResult

public StabilityResult(double meanARI,
 double sdARI,
 int numPairs)

Constructs an object to represent the result of a stability analysis of clustering.

Parameters:

meanARI - The mean Adjusted Rand Index (ARI) across all pairwise comparisons of cluster labelings.

sdARI - The standard deviation of the Adjusted Rand Index (ARI) across pairwise comparisons.

numPairs - The total number of pairwise comparisons considered in the analysis.