A combined PFCM and recurrent neural network-based intrusion detection system for cloud environment. (11th April 2019)

• Record Type: Journal Article
• Authors: Manickam, M.; Ramaraj, N.; Chellappan, C.
• Is Part Of: International journal of business intelligence and data mining. Volume 14:Number 4(2019)
• Page Start: 504
• Page End: 527
• Publication Date: 2019-04-11
• Subjects: cloud computing -- intrusion detection system -- IDS -- possibilistic fuzzy C-means clustering -- PFCM -- recurrent neural network -- RNN; 006.312
• Languages: English
• ISSNs: 1743-8187

Abstract

The main objective of this paper is intrusion detection system for a cloud environment using combined PFCM-RNN. Traditional IDSs are not suitable for cloud environment as network-based IDSs (NIDS) cannot detect encrypted node communication, also host-based IDSs (HIDS) are not able to find the hidden attack trail. The traditional intrusion detection is largely inefficient to be deployed in cloud computing environments due to their openness and specific essence. Accordingly, this proposed work consists of two modules namely clustering module and classification module. In clustering module, the input dataset is grouped into clusters with the use of possibilistic fuzzy C-means clustering (PFCM). In classification module, the centroid from the clusters is given to the recurrent neural network which is used to classify whether the data is intruded or not. For experimental evaluation, we use the benchmark database and the results clearly demonstrate the proposed technique outperformed conventional methods.