public class FMeasure extends java.lang.Object implements ClassificationMeasure
The traditional or balanced Fscore (F1 score) is the harmonic mean of precision and recall, where an F1 score reaches its best value at 1 and worst at 0.
The general formula involves a positive real β so that Fscore measures the effectiveness of retrieval with respect to a user who attaches β times as much importance to recall as precision.
Constructor and Description 

FMeasure()
Constructor of F1 score.

FMeasure(double beta)
Constructor of general Fscore.

Modifier and Type  Method and Description 

double 
measure(int[] truth,
int[] prediction)
Returns an index to measure the quality of classification.

static double 
of(double beta,
int[] truth,
int[] prediction)
Calculates the F1 score.

public FMeasure()
public FMeasure(double beta)
beta
 a positive value such that Fscore measures
the effectiveness of retrieval with respect
to a user who attaches β times as much
importance to recall as precision.public double measure(int[] truth, int[] prediction)
ClassificationMeasure
measure
in interface ClassificationMeasure
truth
 the true class labels.prediction
 the predicted class labels.public static double of(double beta, int[] truth, int[] prediction)
beta
 a positive value such that Fscore measures
the effectiveness of retrieval with respect
to a user who attaches β times as much
importance to recall as precision.