MACHING LEARNING TO DETECT SQL INJECTION

Abstract

SQL Injection vulnerability is being considered as one of the most dangerous security vulnerabilities continuously in the top 10 OWASP. SQL Injection attacks have always seriously affected businesses or private websites and they are still increasing day by day. Due to the increasing demand, the application of machine learning techniques to detect this security breach is to solve the above problem. In this paper, we have used an algorithm called Multinomial Naïve Bayes, K-Nearest Neighbors on the log file in the form of text. As a result, the Multinomial Naïve Bayes algorithm has higher accuracy than the K-Nearest neighbors algorithm in SQL Injection attack.