If we have for two experiments (EXP1 and EXP2), the same accuracy rate and the following results:

EXP1:

a precision of 0.652

F-measure = 0.551

ROC = 0.74

Recall = 0.567

vs.

EXP2:

a precision of 0.521

F-measure = 0.529

ROC = 0.854

Recall = 0.567

As you can see we have different precision rates and ROC areas!

Which of these results is better than the other and why? (The best experience =?)

Can the ROC area be more relevant than precision as a performance measure in a given classification system? Thanks in advance.

More Khadidja Nour el houda Slimani's questions See All
Similar questions and discussions