Precision-Recall

Precision and recall are metrics used to evaluate how well a system identifies relevant items. Precision measures the percentage of items the system labeled as positive that are actually correct—think of it as accuracy among the predicted positives. Recall assesses how many of the true positive items the system successfully identified—like capturing all relevant cases. Together, they help balance the effectiveness of a model: high precision means few false positives, and high recall means few false negatives. These measures are particularly useful in tasks like medical diagnoses or spam detection, guiding improvements in model performance.