When calculating fscore we use the harmonic mean of precision and recall since hm penalizes situations when either of the two metrics is low while the other is large unlike the arithmetic mean. So shouldn't we use this same concept when calculating the mean of k-fold cross validation because we might otherwise get optimistic results based on certain folds with higher values?

        Array = [0.1, 0.2, 0.59, 0.72, 0.96]
        AM = 0.514   HM = 0.261
A more real world example, when predicting alzheimer's disease
        Array = [0.5454, 0.7692, 0.8799, 0.64, 0.8]
        AM = 0.7269   HM = 0.7059


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.