F1 Score is a single metric that combines precision and recall into one number, giving a balanced measure of a model’s accuracy — especially useful when the data is imbalanced.
What it means in plain English
Precision (are the model’s positive predictions correct?) and recall (does it catch all the real positives?) often trade off against each other. The F1 score blends them into one figure, rewarding a model that does well on both. It is particularly valuable when one category is much rarer than another, where plain accuracy can be misleading.
A high F1 score means the model balances catching real cases with avoiding false alarms.
A simple example
For a rare-disease screening model, where positives are uncommon, the F1 score gives a fairer picture than overall accuracy — capturing both how many real cases it catches and how many of its alarms are genuine.
Why it matters
The F1 score is a standard way to evaluate classification models fairly, especially on imbalanced data where accuracy alone can hide poor performance. It condenses two important metrics into one convenient measure of quality.
Related terms
- Precision — one half of the F1 score.
- Confusion Matrix — the source of the numbers behind it.
- Classification — the task F1 evaluates.
Frequently asked questions
What is the F1 score?
It is a single metric that combines precision and recall into one number (their harmonic mean), giving a balanced measure of a classifier’s performance.
When is F1 score useful?
It is especially useful when classes are imbalanced, where plain accuracy can be misleading, because it accounts for both false positives and false negatives.
