Optimizing Features for Intrusion Detection in Cloud Computing using Ensemble Supervised Learning

Abstract

The cloud computing is the phenomenal platform for distributed communication and data sharing. However, the cloud computing platform is vulnerable to many intrusion issues due to their Internet Protocol based loosely coupled distributed communication nature.  The contemporary intrusion defense mechanisms have proven to be suboptimal, which is due to the overwhelmed cloud network traffic and virtual server concepts. Hence, the majority of research works have attracted to the intrusion detection and defense in distributed cloud computing platform. Yet, there is an optimistic scope to define novel intrusion detection and defense mechanisms. Hence, this manuscript portrayed a novel feature selection, optimization, and ensemble classification strategy. The experimental study boosting the ability of the proposal to detect and defend multivariate intrusion strategies, which appears often in cloud computing platform.