Understanding the Performance of Large Language Model to Generate SQL Queries
Recent developments in Artificial Intelligence (AI) have shifted the software development paradigm. Past studies demonstrated how effective AI can generate code for programming purposes. However, to our knowledge, no prior study has been done to evaluate the effectiveness of SQL queries generated by AI. We utilized nine AI assistants to generate SQL queries. Our results reveal that most AI assistants generate inaccurate SQL queries, and based on the results, we provide possible implications for SQL developers.
Authors
- Mohammed Seyam (ORCID: https://orcid.org/0000-0001-6831-3239)
- Minhyuk Ko (ORCID: https://orcid.org/0009-0004-5473-4293)
- Dibyendu Brinto Bose (ORCID: https://orcid.org/0000-0002-1603-3574)
- Wenwen Wang (ORCID: https://orcid.org/0000-0003-0840-4846)
- Chris Brown (ORCID: https://orcid.org/0000-0002-6036-4733)
Institutions
- Virginia Tech (US)
Publication Details
- Published
- 2024-09-02
- DOI
- https://doi.org/10.1109/vl/hcc60511.2024.00048
- Primary Topic
- Data Mining Algorithms and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00