M.A. Thesis:

Prevent attacks in the cloud using intrusion detection systems (IDS)

Nowadays, the use of the Internet and its various programs, such as social networks, has spread rapidly across Different masses of people. At the same time, the amount of data stored on cloud servers is increasing. This provides a good platform for hackers to try to penetrate mobile systems or data storage databases with different methods and try to spy or misuse these data for various purposes. Intrusion detection systems in recent years have been faced with great attention from researchers around the world. Using these systems, they try to identify the attackers at early moments of attacks before they cause serious damage to the system and prevent their activity. The major problem with these intrusion detection systems proposed by various researchers is the low accuracy of these systems in detecting infiltrations. In this thesis, we have tried to propose a new hybrid method using three classification methods (including decision tree, vector support machine and genetic algorithm). The evaluation of the proposed method showed that, compared to other methods developed by other researchers, it has a higher accuracy in detecting infiltrations.

Keywords: Intrusion Detection Systems (IDS), Genetic Algorithm, Support Vector Machine (SVM), Decision Tree.