If the values of Precision and recall are the same then what are the reasons behind that similarity. Also, I want to know what is precision and recall tradeoff and when it occurred?
Dear Habib,
I think it should not necessarily be same.
Dear @Anam Habib
In my opinion, the values of precision and recall can be the same, since they have the same denominator (fp = fn).
precision = tp/tp+fp ||| recall = tp/tp+fn
This means that the algorithm has classified an equal amount of dataset as false positives (fp), as it classified false negatives (fn).
Yes, it is possible.
Precision = TP / (TP + FP)
Recall = TP / (TP + FN)
So, if you have a confusion matrix in which FP = FN (the number of false positives is equal to the number of false negatives), then Precision = Recall
These indicators are in a contradiction - the higher the value of Sensitivity (Recall), the lower the value of Specificity (Precision) and vice versa. In order to optimize algorithm, sometimes use F-measure (
F = 2/(1/precision + 1/recall) ) or the breakeven point (point, where precision = recall).
- Badges
- Science topic