Understand TPR, FPR, Precision and Recall Metrics in Machine Learning – Machine Learning Tutorial

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:

An Introduction to Accuracy, Precision, Recall & F1-Score in Machine Learning – Machine Learning Tutorial

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