Intrusion Detection System based on K-means, Classification and Regression Trees Algorithm

Abstract

The security of computer networks has become a very important aspect in today’s era. An intrusion detection system (IDS) is a type of security software designed to automatically inform administrators when someone is trying to compromise the information system through malicious activities or security policy violations. IDS works by monitoring the system functionalities by checking system vulnerabilities, the integrity of files, and performing analysis of patterns based on known attacks. Intrusion detection is the cycle called distinguishing intrusions. The activity which is entering a framework without consent is called interruption. Interruption Detection Systems are fundamental for security limits. This paper is focused on the analysis of five different techniques like k-means clustering and naive Bayes classification, Enhanced kmeans algorithm, random forest, and weighted k-means, k-means clustering, regression trees algorithm, etc. But some problems are indicated by techniques. These methods analyzed the restrictions of intrusion detection systems by way of low accuracy, high incorrect alarm rate, and time consumption. So, to overcome these problems the proposed method ‘K-means and Classification and Regression Trees Algorithm’ is used to show good accuracy in performance analysis with time complexity by using a hybrid data mining method in the Weka tool. The proposed model of K-means and classification and regression trees algorithm depends on intrusion detection system that gives proper period popular huge information measure of these days’ interruption data set.