Securing Web Applications Against SQL Injection: A Survey on Machine Learning Approaches

Securing Web Applications Against SQL Injection: A Survey on Machine Learning Approaches

1Bhanu Pratap Singh, 2Prof. Manish Kumar Singhal

1M.tech Scholar, 2Associate Professor & H.O.D

1,2Department of Information Technology (IT)

1,2 NRI Institute Of Information Science And Technology, Bhopal (Mp), India,

1   kbpsrmail@gmail.com  2manishsinghal.nirt@gmail.com

Abstract -In this survey paper discuss the SQL injection (SQLi) is one of the most critical and prevalent security vulnerabilities affecting web applications. It occurs when attackers exploit user inputs to inject malicious SQL queries, compromising databases and leading to data breaches, unauthorized access, or even complete control over the system. This survey paper explores the Traditional defense mechanisms, such as input validation and parameterized queries, while effective, often fall short in preventing sophisticated attacks. In recent years, machine learning (ML) has emerged as a promising approach to enhance web application security by detecting and mitigating SQLi threats. This survey explores the state-of-the-art machine learning techniques for securing web applications against SQL injection attacks. We review a wide range of ML-based methods, including supervised, unsupervised, and hybrid approaches. The survey highlights key algorithms such as decision trees, support vector machines (SVM), artificial neural networks (ANN), and deep learning models, emphasizing their effectiveness in identifying SQLi patterns. Additionally, we examine real-time detection systems, anomaly-based solutions, and the challenges associated with implementing ML in real-world web security, such as data scarcity and false positives.

Key Words: SQL Injection, Cross Side Scripting, Denial of Service Attack, Naïve Bias, Gradient Boosting, etc.