Multiple Identity Attack Detection using Correlated Ensemble based Approach

An DDOS is a Multiple Identity Attack utilizing multiple distributed attack sources. Typically, the attackers use a large number of controlled bots distributed in different locations to launch a large number of Multiple Identity Attack. The rapid development of cloud technology in recent years, the attack traffic scale caused by Multiple Identity attacks has been increasing, with the targets including not only business servers, but also internet infrastructures such as firewalls, routers and DNS system as well as cloud bandwidth. The DDOS detection using special mining procedure of neural network is proposed through this literature. This paper describes an approach to handle DDOS in cloud systems. .In the proposed approach correlation between the values are located and the security attributes gives highest correlation and reliability is the next highest correlation values. Both of these attributes serve as root nodes. The comparison between these attributes and training data is made to determine the DDOS. This means complication of calculations is reduced. Classification accuracy of the proposed work is increased considerably..