Building an Efficient of Feature Selection Using Greedy Search Method for HNIDS in Cloud Computing

T. Nathiya and G. Suseendran

The enhancement of HNIDS security prevents the malicious attacks from inside and outside cloud. The intrusion by the malicious attackers in real time cloud networks has major problems for security credentials. The new attacks are identified in cloud networks are not enough to handle the critical situations. The multi-tenant/ small industries are used in private network cloud for storage services. This paper contributes a new feature selection attributes from NSL-KDD datasets. We introduce a new feature selection method, that is identified the network attack packets with minimum attribute selection. The hybrid network intrusion detection system (HNIDS) framework is to detecting the network attacks, while lot of files to store/ from retrieved in cloud for securing the user information. So our proposed greedy search method feature selection using random forest machine learning algorithm to train our framework. Experimental results show the importance of greedy search method/CFS is highly promising accuracy rate of 98.32% and to reduce the false positive rate of the system 0.40% is achieved.

Volume 11 | 04-Special Issue

Pages: 307-316