An Efficient Technique for Detecting SQL Injection Attack in Software Defined Networks using Machine Learning Models

Abstract Due to the large number of connected devices, expanding number of online apps, and growing number of online transactions, data traffic on the Internet is always increasing. Because attackers often target online applications first, web assaults represent a serious danger to organizations. Many solutions, including web application firewalls (WAFs), have been put forward to lessen the impact and severity of these attacks. However, by giving network managers centralized control, software-defined networking (SDN) expertise has greatly enhanced network administration and operation. In this study, we looked at the feasibility of implementing a firewall that can recognize and prevent online assaults using SDN. This research presents a system idea that may concurrently identify SQL injection threats. There isn't a mechanism in place at the moment that can simultaneously identify both threats. This technology guarantees a safe manner to share information and browse the internet. The security of web servers will improve. To anticipate the host that would be struck built on the previous information, we employ well-known machine learning algorithm K-Nearest Neighbours (KNN). Experimental findings demonstrate that Bayesian networks may achieve an average prediction accuracy of 94.56%.