gpt-6-luna
gpt-6-luna masters 110 of the 199 benchmarked knobits of Data modelling: an expected 82% of the competence.
Data modelling, activity by activity
| Activity | Test score | Knobits mastered (expected) |
|---|---|---|
| Capture data requirements | 61% | 59% |
| Define business terms | 66% | 73% |
| Document data rules as SBVR fact types and rule statements | 76% | 87% |
| Build the conceptual model and taxonomies | 93% | 100% |
| Derive the logical data model | 68% | 92% |
| Design the physical data model | 76% | 90% |
| Derive data quality rules | 77% | 89% |
| Generate and review DDL | 62% | 64% |
Where it falls short
The knobits with the lowest probability of mastery: review its output here most carefully.
- K-211 Recognises a definition carrying non-defining content 0%
- K-103 Recognises a statement outside the scope of data requirements 0%
- K-203 Recognises a negative definition 0%
- K-641 Routes a defect to the deliverable responsible for it 1%
- K-104 Recognises an ambiguous business term in a data requirement 9%
- K-133 Recognises a missing or wrong requirement trace 10%
- K-610 Recognises a surrogate key without natural-key uniqueness 11%
- K-309 Recognises a gap in a partition of cases 15%
- K-828 Recognises a business date stored with a time of day 19%
- K-821 Recognises a restricted column without access design 20%
- K-810 Recognises a truncated identifier taken as the real name 20%
- K-117 Recognises a missing or wrong requirement type 20%
- K-822 Recognises greenfield DDL aimed at a populated database 25%
- K-823 Recognises a drop script used as a rollback 25%
- K-201 Recognises a definition circular within itself 27%