cc.mallet.cluster.neighbor_evaluator
Interface NeighborEvaluator

All Known Implementing Classes:

ClassifyingNeighborEvaluator, MedoidEvaluator, PairwiseEvaluator, RandomEvaluator, RankingNeighborEvaluator

public interface NeighborEvaluator

Scores the value of changing the current Clustering to the modified Clustering specified in a Neighbor object. A common implementation of this interface uses a Classifier to assign a score to a Neighbor.

Since:

1.0

Version:

1.0

Author:

"Aron Culotta"

Method Summary

double	evaluate(Neighbor neighbor)
double[]	evaluate(Neighbor[] neighbors)
void	reset() Reset the state of the evaluator.

Method Detail

evaluate

double evaluate(Neighbor neighbor)

Parameters:

neighbor -

Returns:

A higher score indicates that the modified Clustering is preferred.

evaluate

double[] evaluate(Neighbor[] neighbors)

Parameters:

neighbors -

Returns:

One score per neighbor. A higher score indicates that the modified Clustering is preferred.

reset

void reset()

Reset the state of the evaluator.