Neural network based intrusion detection using Bayesian with PCA and KPCA feature extraction
Tareek M. Pattewar, Harshal A. Sonawane · 2015
In today's digital world use of computer systems, computer networks and internet are increasing rapidly. Due to this, information is processed digitally. So, providing effective security to such digital data is crucial task. There are some tools and systems are available for the security of digital information. From these tools and systems intrusion detection system is an important method. It is necessary to design good intrusion detection system which will give good accuracy rate, less error rate, which will take minimum time and memory in the performance of the system. This paper presents the two techniques of intrusion detection system. Working of these techniques are based on neural network. For feature extraction in first technique, we are using principal component analysis (PCA) method. In second technique, we are using Bayesian for noise reduction and kernel principal component analysis (KPCA) for feature extraction. These both techniques are promising for giving good performance results. Systems are implemented using java technology.