Hybrid Intrusion Detection System Using K-Means and Classification and Regression Trees Algorithms

Intrusion detection is the process called indentifying invasions. The action to enter a system without permission is called intrusion. By adding advanced technologies to mobile phones, such as smart phones, tablets, smart devices, other computing devices, the number of Internet users are increasingly growing. Therefore, network security is very important for all Internet users. IDS are essential for security limits. So, now Internet consumers are considered mandatory safety devices for critical networks. There are many traditional techniques of intrusion detection. In research on traditional intrusion detection technology analysis, the statistical model for setting up rules, management and aggression capability and so on are still some disadvantages and disabilities, because the actual test results cannot meet the requirements. There are many current methods used in. Each method has advantage and disadvantage. Intrusion detection can also be considered as a classification problem. In this research we use K-means algorithm and classification and regression trees (CART) algorithm. The purpose of this paper is to show good accuracy in performance analysis with time complexity by using hybrid data mining method. This model is verified by KDD'99 data set.