A better Beta for the H measure of classification performance

Details A better Beta for the H measure of classification performance A better Beta for the H measure of classification performance

2012-02-12

arxiv.org/abs/1202.2564v1

Statistics

Methodology

Preprint. Keywords: supervised classification, classifier performance, AUC, ROC curve, H measure

Scientific paper

The area under the ROC curve is widely used as a measure of performance of classification rules. However, it has recently been shown that the measure is fundamentally incoherent, in the sense that it treats the relative severities of misclassifications differently when different classifiers are used. To overcome this, Hand (2009) proposed the $H$ measure, which allows a given researcher to fix the distribution of relative severities to a classifier-independent setting on a given problem. This note extends the discussion, and proposes a modified standard distribution for the $H$ measure, which better matches the requirements of researchers, in particular those faced with heavily unbalanced datasets, the $Beta(\pi_1+1,\pi_0+1)$ distribution.

Affiliated with

Also associated with

No associations

Rating

A better Beta for the H measure of classification performance does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with A better Beta for the H measure of classification performance, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A better Beta for the H measure of classification performance will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-680310