claude-haiku-5-5
claude-haiku-5-5 masters 148 of the 199 benchmarked knobits of Data modelling: an expected 86% of the competence.
Data modelling, activity by activity
| Activity | Test score | Knobits mastered (expected) |
|---|---|---|
| Capture data requirements | 74% | 70% |
| Define business terms | 50% | 82% |
| Document data rules as SBVR fact types and rule statements | 88% | 93% |
| Build the conceptual model and taxonomies | 95% | 100% |
| Derive the logical data model | 83% | 94% |
| Design the physical data model | 78% | 92% |
| Derive data quality rules | 84% | 95% |
| Generate and review DDL | 67% | 64% |
Where it falls short
The knobits with the lowest probability of mastery: review its output here most carefully.
- K-606 Recognises rolled-up subtypes losing conditional constraints 0%
- K-104 Recognises an ambiguous business term in a data requirement 0%
- K-204 Recognises a definition too broad for its concept 0%
- K-103 Recognises a statement outside the scope of data requirements 0%
- K-816 Recognises one decimal scale for every currency 0%
- K-309 Recognises a gap in a partition of cases 1%
- K-803 Recognises an object in the SQL missing from the model 6%
- K-211 Recognises a definition carrying non-defining content 10%
- K-209 Recognises a hidden definition of a second concept 10%
- K-828 Recognises a business date stored with a time of day 14%
- K-117 Recognises a missing or wrong requirement type 19%
- K-821 Recognises a restricted column without access design 20%
- K-810 Recognises a truncated identifier taken as the real name 21%
- K-718 Recognises a misfiled data quality dimension 23%
- K-822 Recognises greenfield DDL aimed at a populated database 27%