ML - F1 score

The F1 score is a metric used to evaluate the performance of a classification model, especially when the classes are imbalanced.

It is the harmonic mean of two other metrics:

• Precision: Of the items the model predicted as positive, how many were actually positive?

• Recall: Of all actual positive items, how many did the model correctly identify?

Formula

Intuition

• High precision but low recall → F1 is low

• High recall but low precision → F1 is low

• F1 is high only when both precision and recall are high

When to use it

• When you need a single score that balances precision and recall

• When the dataset is imbalanced and accuracy can be misleading

• Common in spam detection, medical diagnosis, fraud detection, etc.