Machine Learning

Confusion Matrix Metrics

The set of performance metrics derived from the confusion matrix including true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN).

Why It Matters

Understanding TP, TN, FP, and FN is essential for evaluating any classification model. Different applications require prioritizing different metrics.

Example

In a COVID test: TP = correctly detected infection, FP = healthy person told they are infected, FN = infected person told they are healthy, TN = correctly identified healthy.

Think of it like...

Like the four possible outcomes of a fire alarm: it correctly sounds for a real fire (TP), falsely sounds with no fire (FP), fails to sound during a fire (FN), or correctly stays silent (TN).

Related Terms