In simple words: Accuracy alone can hide important mistakes.
Accuracy
Accuracy is correct predictions divided by all predictions. If 85 of 100 predictions are correct, accuracy is 85%. But if 99 of 100 payments are genuine, a model that says “genuine” every time gets 99% accuracy while catching no fraud.
Precision and recall
Precision asks: of the cases predicted positive, how many were truly positive? Recall asks: of all real positives, how many did we find? F1 score balances precision and recall using their harmonic mean. Choose the measure according to the cost of each error.
See the flow
predictions + true answers→metrics
Simple example
A fraud alert system with high recall catches most fraud. Low precision means many genuine payments are also blocked.
Remember: Accuracy alone can hide important mistakes.