Package cc.mallet.cluster.evaluate

Class ClusteringEvaluator

java.lang.Object

cc.mallet.cluster.evaluate.ClusteringEvaluator

Direct Known Subclasses:

AccuracyEvaluator, BCubedEvaluator, ClusteringEvaluators, MUCEvaluator, PairF1Evaluator

public abstract class ClusteringEvaluator
extends java.lang.Object

Evaluates a predicted Clustering against a true Clustering.

Since:

1.0

Version:

1.0

Author:

"Aron Culotta"

Constructor Summary

Constructors

Constructor	Description
ClusteringEvaluator()

Method Summary

Modifier and Type	Method	Description
java.lang.String	evaluate(Clustering[] truth, Clusterer clusterer)
java.lang.String	evaluate(Clustering[] truth, Clustering[] predicted)
abstract java.lang.String	evaluate(Clustering truth, Clustering predicted)
abstract java.lang.String	evaluateTotals()
abstract double[]	getEvaluationScores(Clustering truth, Clustering predicted)

Methods inherited from class java.lang.Object

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

Constructor Detail

ClusteringEvaluator

public ClusteringEvaluator()

Method Detail

evaluate

public abstract java.lang.String evaluate(Clustering truth,
                                          Clustering predicted)

Parameters:

truth -

predicted -

Returns:

A String summarizing the evaluation metric.

evaluate

public java.lang.String evaluate(Clustering[] truth,
                                 Clustering[] predicted)

evaluate

public java.lang.String evaluate(Clustering[] truth,
                                 Clusterer clusterer)

getEvaluationScores

public abstract double[] getEvaluationScores(Clustering truth,
                                             Clustering predicted)

evaluateTotals

public abstract java.lang.String evaluateTotals()

Returns:

If the ClusteringEvaluator maintains state between calls to evaluate, this method will return the total evaluation metric since the first evaluation.