Precision
Definition
Precision shows what proportion of positive hits are actually correct.
The proportion of truly relevant or positive cases among all cases predicted as positive.
Simply explained
Of everything marked as suitable or positive, the actually correct proportion is 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
Of 100 parts marked as defective, 90 are actually defective; precision is 0.9.
Limits and pitfalls
- Confusing with accuracy or recall and ignoring class imbalance.
Origin & history
Precision is a classic metric for classification and information retrieval.
Use in mid-sized companies
- Possible application: In quality control, determine what proportion of parts marked as defective are actually faulty.
Frequently asked questions
What does precision mean?
The proportion of truly relevant or positive cases among all cases predicted as positive. What mistakes should be avoided? Confusing with accuracy or recall and ignoring class imbalance.
Sources
Source checked on 10.10.2026. Editorial application examples are hypothetical; no proof of real customer results.
Editorially reviewed on: