Improve the performance of support vector machine (SVM) in design of an intelligent IDS using feature ranking

with the expansion of the activities computer networks users in around the world and followed by growth complexity and size of data, detecting unauthorized activity and abuse is very difficult. Hence the computer network security is of great importance. So in this paper, intelligent security system based on machine learning for intrusion detection is designed. Support vector machine one of the intelligent analysis techniques in intrusion detection has m any usages that in this paper are being tried with SVM algorithm will classify authorized and unauthorized activities in computer networks. Finally, a new method scilicet using features ranking are propone that SVM performance is achieved improve the minimum false positive rate and precision rate in classification.

کلیدواژه ها:

machine learning ، intrusion detection ، support vector machine ، feature selection ، Entropy measure