SSQLi: A Black-Box Adversarial Attack Method for SQL Injection Based on Reinforcement Learning

SQL injection is a highly detrimental web attack technique that can result in significant data leakage and compromise system integrity. To counteract the harm caused by such attacks, researchers have devoted much attention to examination of detection techniques, which progressed from traditional signature-based methods machine- deep-learning-based models. These techniques demonstrated promising results on existing datasets; however, most studies overlooked impact adversarial particularly black-box methods. This study addressed shortcomings current proposed reinforcement-learning-based method. The proposal included an innovative vector transformation approach for original payload, comprehensive attack-rule matrix, method adaptive generation examples. Our was evaluated application firewalls (WAF) models based deep-learning methods, generated examples successfully bypassed at rate up 97.39%. Furthermore, there substantial decrease accuracy model after multiple attacks had been carried out via