In machine learning, we may see these metrics: TPR, FPR, Precision and Recall metrics. In this tutorial, we will introduce their relation.
If our result is:
Precision and Recall
They are computed as follows:
We can understand Precision and Recall as follows:
TPR and FPR
TPR = True Positive Rate
FPR = False Positive Rate
They can be computed as:
TPR = TP / (TP + FN)
FPR = FP / (FP + TN)
They are:
Comparing TPR, FPR and Precision, Recall, we can find:
TPR = Recall
FNR = 1-TPR