A technique utilizing Machine Learning and Convolutional Neural Networks (CNN) for the identification of SQL Injection Attacks

Analyzing logs can aid in debugging or optimizing system performance. A comprehension of system efficacy is frequently linked to an awareness of the manner in which the system’s resources are utilized. SQL injection attacks, specifically targeting web applications, have become a significant risk in the realm of cybersecurity. SQL injection attacks primarily result in the unauthorized disclosure of user data, allowing for data manipulation, updating, and deletion within web applications. Conventional methods employed to mitigate SQL injections encompass rule-based matching and other associated techniques that have a narrow scope in addressing only a few types of SQL injections. This research study examines code specifically designed to identify and prevent SQL injection attacks. Various supervised Machine Learning techniques and Convolutional Neural Network (CNN) models are employed to assess the model. The CNN model that was suggested exhibited a notable accuracy rate of 91% and outperformed other machine learning algorithms. Furthermore, the report gives a comprehensive analysis of several machine learning algorithms employed for the detection of SQL injection attacks. The study evaluates the effectiveness of different methods in detecting SQL injection attacks using F1 Score, accuracy, recall and precision metrics. Also compares the performance of machine learning and CNN models.