The score is a performance metric for binary classification tasks. It is often compared with accuracy because it can accommodate class imbalance. As it focuses on positive cases, it is less useful in situations where false negatives are particularly costly. However, it is an excellent metric for information retrieval because it incorporates information about both precision and recall. Recall that and The score is the harmonic mean of these quantities: Using the abbreviations , , , to refer to the quantities in the binary confusion matrix, we can derive as follows. The score can be generalized as the F-beta score (“f-score”) (also known as the -score). Evaluation (performance) metrics