A Method of SQL Injection Attack Detection Based on Large Language Models

A Method of SQL Injection Attack Detection Based on Large Language Models

연구 분야: Databases

학회: 2024 2nd International Conference on Computer Network Technology and Electronic and Information Engineering (CNTEIE)

SQL injection attacks are a serious threat to the security of cyberspace. In view of the problems with traditional SQL injection attack detection methods, such as high false positive rates and insufficient generalization, this paper proposes a new detection method based on large language models (LLMs). Through prompt engineering and instruction tuning technologies, a dedicated large language model for SQL injection attack detection is constructed. The experimental results show that the proposed method achieves an accuracy of 99.85% and a false positive rate of 0.26% on general benchmark datasets, which verifies the effectiveness of large language models in the field of SQL injection attack detection.