Efficient Intrusion Detection Using Machine Learning Approaches

Abstract

The fast propagation of computer networks has changed the viewpoint of newlinenetwork security. An easy accessibility conditions cause computer network as newlinesusceptible against several threats from hackers. Threats to networks are newlinenumerous and potentially devastating. Up to the moment, researchers have newlinedeveloped Intrusion Detection Systems (IDS) capable of detecting attacks in newlineseveral available environments. newlineA boundlessness of methods for misuse detection as well as anomaly newlinedetection has been applied. Many of the technologies proposed are newlinecomplementary to each other, since for different kind of environments some newlineapproaches perform better than others. This research presents a new intrusion newlinedetection system that is then used to survey and classify them. The taxonomy newlineconsists of the detection principle, and second of certain operational aspects of newlinethe intrusion detection system. newlineIn our research we have used algorithms like Random Forest (RF), newlineSupport Vector Machine (SVM), Lexicographic Game Method, Artificial newlineNeural Network (ANN), Enhanced Convolution Neural Network (ECNN), Collaborative Game Method and Petri Net Process. All are measured in terms of newlineaccuracy.