edu.uci.jforests.eval.ranking

Class RankingEvaluationMetric

java.lang.Object

edu.uci.jforests.eval.EvaluationMetric

edu.uci.jforests.eval.ranking.RankingEvaluationMetric

Direct Known Subclasses:

MAPEval, NDCGEval, PrecisionEval, URiskAwareEval

public abstract class RankingEvaluationMetric
extends EvaluationMetric

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	RankingEvaluationMetric.SwapScorer

Constructor Summary

Constructors

Constructor and Description
RankingEvaluationMetric(boolean isLargerBetter)

Method Summary

Modifier and Type	Method and Description
static double[]	computeNaturalOrderScores(int numTargets, int[] boundaries) This gives the ordering of docuuments if they remained in their original order
EvaluationMetric	getParentMetric()
abstract RankingEvaluationMetric.SwapScorer	getSwapScorer(double[] targets, int[] boundaries, int trunc, int[][] labelCounts)
double	measure(double[] predictions, Sample sample)
abstract double[]	measureByQuery(double[] predictions, Sample sample)

Methods inherited from class edu.uci.jforests.eval.EvaluationMetric

isFirstBetter, largerIsBetter

Methods inherited from class java.lang.Object

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

Constructor Detail

RankingEvaluationMetric

public RankingEvaluationMetric(boolean isLargerBetter)

Method Detail

computeNaturalOrderScores

public static double[] computeNaturalOrderScores(int numTargets,
                                                 int[] boundaries)

This gives the ordering of docuuments if they remained in their original order

Parameters:

numTargets - how many documents we want in the ranking

boundaries - the query offsets

Returns:

an array with the ranking offsets for each query

getSwapScorer

public abstract RankingEvaluationMetric.SwapScorer getSwapScorer(double[] targets,
                                                                 int[] boundaries,
                                                                 int trunc,
                                                                 int[][] labelCounts)
                                                          throws java.lang.Exception

Parameters:

targets - Targets of all documents in dataset. Equals Dataset.*.length

boundaries - Start and end indices of each query within the dataset.

trunc - Truncation level for evaluation measure

labelCounts - How many of each label value

Throws:

java.lang.Exception - if things go pear shaped

measureByQuery

public abstract double[] measureByQuery(double[] predictions,
                                        Sample sample)
                                 throws java.lang.Exception

Throws:

java.lang.Exception

measure

public double measure(double[] predictions,
                      Sample sample)
               throws java.lang.Exception

Specified by:

measure in class EvaluationMetric

Throws:

java.lang.Exception

getParentMetric

public EvaluationMetric getParentMetric()