A Machine Learning Approach to SQL Injection Detection in Web Applications

Abstract

Web Applications have become integral to modern life, storing vast amounts of sensitive data and thus becoming prime targets for attackers exploiting vulnerabilities like SQL injection. This paper underscores the importance of data pro-tection, particularly regarding SQL injection detection in web applications. Both traditional and advanced detection strategies are looked at with a focus on deep learning and machine learning methods. In this paper, machine learning algorithms for SQL attack detection are explored with a focus on necessary aspects like dataset selection, feature extraction, and model evaluation. The Methodology section outlines the process for conducting our experiments, including data collection, preparation, and model training. By assessing multiple datasets and algorithms, the paper demonstrates the effectiveness of machine learning models like Random Forest and LSTM in achieving high accuracy and F1 scores. In summary, the study emphasizes the importance of protecting web applications from SQL injection attacks. By leveraging machine learning algorithms, web applications can mitigate risks, protect user data, and foster a secure digital environment.