Author name: Jamal Hussein

A practical guide to classification metrics — accuracy, precision, recall, F1 and ROC-AUC — with a decision process for choosing the right one for balanced vs imbalanced data.

The ROC curve plots true positive rate vs false positive rate across all thresholds; AUC summarises it as one number. Learn how to read them, what a good AUC is, and ROC-AUC vs PR-AUC.

The F1 score is the harmonic mean of precision and recall: F1 = 2·(P·R)/(P+R). Learn why it uses a harmonic mean, when to use it, F-beta, and macro vs micro vs weighted F1.

Precision is TP/(TP+FP) and recall is TP/(TP+FN). Learn the difference, the precision–recall trade-off, and which to prioritise for imbalanced problems like fraud, disease and spam.

A confusion matrix compares predicted vs actual labels across four cells — TP, FP, TN, FN — and every classification metric (accuracy, precision, recall, F1) is derived from them.