Train an Intelligent Database with ChatGPT
The integration of AI into database management has the potential to reduce the complexities of SQL, making databases more accessible to non-technical users. Traditional database systems often require users to possess specialized knowledge in SQL, which creates significant barriers for non-technical professionals who need to query or manipulate data. This issue is particularly evident in industries such as marketing and finance, where professionals are often required to interact with databases but lack the necessary coding expertise. This research focuses on developing a Java-based system that utilizes ChatGPT to convert natural language into SQL commands, allowing non-technical users to interact with databases in a more intuitive and user-friendly manner. The primary research question addresses how AI, specifically ChatGPT, can bridge the gap between natural language and complex SQL syntax, enabling seamless database interactions without requiring users to learn SQL. The methodology involves configuring a MySQL environment, integrating the OpenAI API, and iteratively refining the system to handle basic database operations such as querying, inserting, and deleting data. The data for this study is sourced from a pre-existing MySQL database, which includes various sample datasets commonly used for testing database operations. The database contains multiple tables with different types of data, allowing the system to perform a variety of SQL operations. The results show that this system successfully lowers the barriers to database access, significantly improving the ease of use for non-technical users. However, limitations remain in handling more complex SQL queries and occasional inaccuracies in generated SQL commands. This paper concludes that AI-driven tools, like ChatGPT, have great potential to enhance database usability. Future work will focus on improving the accuracy of generated SQL commands, expanding the system's ability to handle complex queries, and refining the user interface to improve user experience.
Authors
- Yifeng Jiang (ORCID: https://orcid.org/0009-0005-4398-3611)
Institutions
- Xi’an Jiaotong-Liverpool University (CN)
Publication Details
- Published
- 2024-12-27
- DOI
- https://doi.org/10.1145/3718491.3718667
- Primary Topic
- COVID-19 diagnosis using AI
- Type
- article
- Field-Weighted Citation Impact
- 0.00