APPLICATION OF ARTIFICIAL INTELLIGENCE TO DETECT SQL INJECTIONS

Abstract

The development of information technologies has significantly affected people's lives. At the same time, the importance of information security is becoming more and more significant. Protecting the integrity, accessibility, and confidentiality of users' personal data in the database is the main goal of information security. One of the biggest threats to the security of web applications is SQL injection. SQL injection refers to a type of attack that allows bypassing access control to web applications for direct database management, jeopardizing the application's information security. In this article, we consider the identification of SQL injections using artificial intelligence methods. For this task, machine learning and neural network models were developed using the following algorithms: Naive Bayesian Classifier, Support Vector Machine, Logistic Regression, Decision Tree, Random Forest, XGBoost, AdaBoost, and Deep Neural Networks. To assess the effectiveness of data classification using the developed models, the indicators of accuracy, precision, recall, and F1-score were used, which reached values in the range of 0.95 – 0.99.