Precision at k
Definition
Precision@k measures the relevant proportion among the top search results.
The proportion of relevant results among the top k positions of a ranked list.
Simply explained
Only the first fixed positions are considered and how many of them really fit are counted.
Why it matters in practice
The term helps to select, develop, evaluate AI systems in a well-founded manner and realistically assess their technical risks.
Example from business
Among the first ten search results, eight are relevant; P@10 is 0.8.
Limits and pitfalls
- Do not omit specifying k or treat missing relevance judgments as irrelevant.
Origin & history
P@k is an established metric in information retrieval and recommendation systems.
Use in mid-sized companies
- Possible example: Test a company search with the same test queries to see how many of its first ten hits are relevant.
Frequently asked questions
What does Precision among the first k hits mean?
The proportion of relevant results among the top k positions of a ranked list. What mistakes should be avoided? Do not omit specifying k or treat missing relevance judgments as irrelevant.
Sources
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: