A Machine Learning Based Intrusion Impact Analysis Scheme for Clouds

Abstract

Clouds represent a major paradigm shift, inspiring the contemporary approach to computing. They present fascinating opportunities to address dynamic user requirements with the provision of on demand expandable computing infrastructures. However, Clouds introduce novel security challenges which need to be addressed to facilitate widespread adoption. This paper is focused on one such challenge - intrusion impact analysis. In particular, we highlight the significance of intrusion impact analysis for the overall security of Clouds. Additionally, we present a machine learning based scheme to address this challenge in accordance with the specific requirements of Clouds for intrusion impact analysis. We also present rigorous evaluation performed to assess the effectiveness and feasibility of the proposed method to address this challenge for Clouds. The evaluation results demonstrate high degree of effectiveness to correctly determine the impact of an intrusion along with significant reduction with respect to the intrusion response time.

Keywords

• Cloud computing
• Cloud security
• Intrusion severity analysis
• Intrusion Detection
• Intrusion Response.