Research on System of Genetic Algorithm-Based SQL Injection Test Data

With the rapid development of network science and technology, an increasing number of network security vulnerabilities have emerged, among which SQL injection is a particularly severe security vulnerability. To effectively detect vulnerabilities, how to construct a high-quality test dataset has become a research hotspot. This paper proposes and implements a system design based on genetic algorithms to optimize SQL injection test data, which can automate the generation of test data. Specifically, it involves filtering, completion, and feature extraction of SQL injection attack characteristics using filtering rules and completion rules. The system also utilizes genetic algorithms to process the test dataset through selection, crossover, mutation, and other operations in multiple iterations to generate the best SQL injection test dataset. Experimental results demonstrate that this design can generate high-quality test datasets, effectively discover SQL injection vulnerabilities, and improve the comprehensiveness and accuracy of vulnerability detection. It addresses the drawbacks of high cost in manually constructing test data and low quality of test data generated based on rules.