UTILIZING VOTING CLASSIFIER FOR NETWORK INTRUSION DETECTION

The intrusion detection system helps in monitor network traffic and identify various attack vectors and accordingly mitigate them in a timely manner, as the use of Internet has increased exponentially, the Intrusion detection system has become a vital part of Information technology infrastructure. As more and more users are connected via Internet, these users are vulnerable to various forms to security attacks which can lead to data theft, fraudulent transactions, privacy breach etc. The existing Intrusion Detection System (IDS) provide low detection accuracy, which might lead to either false positives or false negative. In this research various types of machine learning algorithms are applied to get the best results in prediction of Network Intrusion and also by using voting Techniques. Classifiers like SVM, K Means, Naive Bayes, voting classifier are used and the result analyzed. The result of various models is then compared to find the best method for detection.