Review on intrusion detection using feature selection with machine learning techniques

The Network Security is the most exciting duty in the modern world for development of internet applications, moreover numerous attacks also increased in internet and hence it is need to developing novel techniques for attack detection and prevention more efficiently. It can be achieved by proposing new algorithms for intrusion detection also. To prevent these by Intrusion Detection System (IDS), Access control, and key management. Among these, IDS is more popular for ensuring network security. The present works focus on artificial intelligence based approaches for developing framework of Intrusion Detection to identify the undesirable attacks and to prevent unauthorized access. This paper is a general study of the existing system is surveyed with bench mark data set to identify the unusual attacks and also for understanding the current issues in problems of intrusion detection. The complete studies are mentioned here for detecting different types of attacks and its relevant issues using the machine learning classification algorithms. Furthermore, this investigation evaluates the Performance analysis of the existing IDS using selecting attribute feature and classification of machine learning methods. Finally, experimental results obtained in various methods, we recommend few superior features selection and machine learning algorithm find out the efficient algorithm that can be used for particular intrusive attack detection. (C) 2019 Elsevier Ltd. All rights reserved.