Table 7 Comparison of scores when GitHub issues are included as background knowledge
From: Dynamic Retrieval Augmented Generation of Ontologies using Artificial Intelligence (DRAGON-AI)
Method | Model name | Accuracy | Score | Consistency |
|---|---|---|---|---|
DRAGON | gpt-3.5-turbo | 4.067 | 3.626 | 3.709 |
DRAGON + gh | gpt-3.5-turbo | 4.182 | 3.717 | 3.733 |
DRAGON | gpt-4 | 4.041 | 3.608 | 3.754 |
DRAGON + gh | gpt-4 | 4.241 | 3.805 | 3.893 |
curator | human | 4.439 | 4.158 | 4.182 |