F1 Score

Definition

The harmonic mean of Precision and Recall. It provides a single, balanced metric to evaluate a model when you need to balance both False Positives and False Negatives, particularly on uneven datasets.

Explain Like I'm New

The ultimate combined grade. Since Precision and Recall are a see-saw (if one goes up, the other usually goes down), the F1 Score punishes extreme imbalances. To get a high F1 score, the AI must be both highly Precise AND have high Recall.

Real World Example

A data scientist trains 5 different ML models to detect credit card fraud. Since most transactions are legit, 'Accuracy' is a useless metric. The scientist compares the models using their F1 Scores to find the best all-around performer.

Common Use Cases

•Imbalanced datasets
•Model benchmarking

Interview Questions

basic

If a model has 100% Recall but 10% Precision, will its F1 Score be high or low?

intermediate

Why does the F1 score use a 'Harmonic Mean' instead of a simple average?

Flash Cards

Question

High or low?

Click to reveal answer

Answer

Low. The F1 Score severely penalizes models that sacrifice one metric entirely for the other.

Question

Why harmonic mean?

Click to reveal answer

Answer

A simple average of (1.0 and 0.0) is 0.5. But an AI with 0 Precision is completely useless. The Harmonic Mean equation ensures that if either Precision OR Recall drops to 0, the final F1 score drops all the way to 0, accurately reflecting the model's failure.