ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Feasibility of Machine Learning Techniques to Reduce False Alarm in Intrusion Detection?

Journal: International Journal of Computer Science and Mobile Computing - IJCSMC (Vol.2, No. 13)

Publication Date:

Authors : ;

Page : 166-172

Keywords : Artificial Neural Network; Adaptive Learner for Alert Classification; K Nearest Neighbour; Data Mining; Genetic Algorithm; Particle Swarm Optimization; Multilayer Perception; Detection Rate; False Positive; Support Vector Machine;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

The sophisticated recent advances in networking increased the dependability of humans in Network technology everyday life. Now-a-days the number of attacks on networks got increased drastically. With the development of large open networks, security threats have increased significantly in the past two decades. So mitigating those attacks is one of the significant interests of researchers in the network security. The goal of an intrusion detection system is to provide a wall of defence to confront the attacks of computer systems on internet. Machine Learning algorithms have been successfully applied to intrusion detection; however the true positive and false positive trade-offs is always a major challenge in the choice of the algorithms. This paper explores the wide variety of machine learning algorithms focusing on the feasibility of these algorithms for the purpose of reducing false positives.

Last modified: 2014-01-11 01:24:28