Insider Attack Detection Using Deep Belief Neural Network in Cloud Computing

Cloud computing is a high network infrastructure where users, owners, third authorized and customers can access store their information quickly. The use of cloud has realized the rapid increase in every field need for centralized location processing efficiently. This nowadays highly affected by internal threats user. Sensitive applications such as banking, hospital, business are more likely real user threats. An intruder presented set member network. After becoming an insider network, they will try to attack or steal sensitive data during sharing conversation. major issue today's technological development identifying threat When lost, compromising users difficult. Privacy security not ensured, then, usage trusted. Several solutions available external However, be addressed. In this research work, we focus on solution using artificial intelligence technique. possible nodes weak users’ systems. They log id, connect pretend trusted node. Then, easily hack insider, them very These types attacks intelligent solutions. A machine learning approach widely used issues. To date, existing lags classify attackers accurately. hijacking process absurd, which motivates young researchers provide our proposed track interaction behavior pattern deep mouse movements clicks keystrokes stored database. belief neural designed restricted Boltzmann (RBM) so that layer RBM communicates with previous subsequent layers. result evaluated Cooja simulator based environment. accuracy F-measure improved compared when long short-term memory support vector machine.