Analyses of SQL injection Attacks and Their detection techniques: A Systematic Literature Review

Abstract

SQL injection attacks are a common threat to web applications, exploiting weaknesses in database access mechanisms to allow unauthorized execution of SQL queries by attackers. SQL, a programming language for managing relational databases, enables users to interact with databases by adding, deleting, and updating data. Such attacks may be serious security risks to web applications, as they can lead to unauthorized access and retrieval of restricted data stored in databases [2]. This research paper a review on comparable analysis to find out more on different types of attacks, and their detection techniques.Various detection techniques, including machine learning algorithms and deep neural networks, have been proposed to combat SQL injection attacks. Deep neural networks, a subset of machine learning, are adept at understanding patterns and representations in extensive datasets, making them effective in understanding diverse data types like text, images, and sounds. However, one drawback of neural networks is their tendency to make overly confident predictions, especially when faced with unfamiliar data beyond their training scope. Researchers have introduced methods such as deep ensembles and Bayesian neural networks to estimate predictive uncertainty and mitigate overconfidence. Despite the potential of neural networks in various applications, it's essential to address their inclination towards overconfident predictions and potential drawbacks like overfitting. techniques like deep ensembles can enhance the reliability of neural network predictions by estimating predictive uncertainty effectively. Moreover, deep learning has demonstrated successful applications in the realm of Web security detection, showcasing its potential impact across a wide range of domains [4].