An interesting performance measure that Weka gives is the Weighted average of TP rate, FP rate, Precision, Recall, F-measure, ROC area and so on. But this can be confusing at some times, especially for weighted average F-measure, which is not actually the harmonic mean of the weighted average precision and recall. How does then Weka calculate the weighted average F-measure (and others)? This tutorial clearly and quantitatively answers that!
آی-ویدئو