Confusion Matrix Explained

A confusion matrix is a table that shows how a classification model performed by breaking down its predictions into correct and incorrect categories.

It compares actual labels with predicted labels, showing four possible outcomes: true positives, true negatives, false positives, and false negatives.

A true positive means the model correctly predicted the positive class, while a false positive means it predicted positive when the actual result was negative. The same logic applies in reverse for true negatives and false negatives.

Reading a confusion matrix helps you understand not just how often a model is right, but exactly what kind of mistakes it tends to make, which a single accuracy score cannot show on its own.