Measuring the Performance of a Classifier

This program is provided to support the description of the calculation of performance measures given in Chapter 12 of Principles of Data Mining (second edition) by Max Bramer.

Given a table (such as Figure 12.1 on Page 176) containing the number of true and false positives and negatives produced by a classifier, it calculates the performance measures shown in Figure 12.2 (Page 179).


Enter the values of TP, FP, FN and TN and click on 'Submit'

 
Predicted Class
Positive
Negative
Actual Class
Positive
TP
FN
Negative
FP
TN

TP - number of true positives: instances classified as positive that are actually positive
FP - number of false positives: instances classified as positive that are actually negative ('type 1' errors)
FN - number of false negatives: instances classified as negative that are actually positive ('type 2' errors)
TN - number of true negatives: instances classified as negative that are actually negative