SYN Flood Attack From TCP/IP Using Statistical Analysis of Machine Learning Techniques

Abstract

Most of the network management affected by SYN flood attack needed to secure the server since being harassed by malevolent attackers. The Transmission control protocol synchronize (SYN) flood malicious attack happens whilst the hacker floods the network system with requests in order to overcome the intention and make it unable to respond to new real connection requests. This makes many of the intention server’s communications ports into half-open state. If this kind of attack established in big data analysis, Artificial Intelligence, as well as Internet of Things, then SYN flood has to be found. So we proposed novel model for detecting the attack through statistical analysis in machine learning techniques such as decision tree classifier, ensemble gradient boosting classifier, MLP classifier for enhancing the model performance by evaluating the metrics such as accuracy, F-Score, FNR.