Precision at k

Precision at k (P@k) measures how many of the top k items recommended or retrieved are relevant to the user’s interests. For example, if a search engine shows 10 results and 7 are useful, the precision at 10 is 70%. It's a way to evaluate how well a system ranks relevant items near the top of its list. Higher precision at k indicates more relevant items are prioritized earlier, improving the user’s experience by reducing the need to sift through less useful results.