Behavior Reliance Anomaly Detection with Customized Compact Prediction Trees
K. Venkateswara Rao1, T. Uma Devi2

1K. Venkateswara Rao, Ph.D Scholar at GITAM (Deemed to be University), Visakhapatnam, India and Associate Professor (CSE), Chebrolu Engineering College, Chebrolu, (A.P.) India.
2Dr. T.UmaDevi, Associate Professor in the Department of Computer Science, GITAM (Deemed to be University), Visakhapatnam, (A.P.) India.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 1495-1502 | Volume-8 Issue-8, June 2019 | Retrieval Number: H6640068819/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Since beginning of the cloud computing, service providers are always concentrated to protect their cloud technical assets from external adversaries and neglected the insider threat mitigation activities, caused to remain the today clouds as vulnerable to malicious insider threats. Malicious insiders are the integral part of our cloud environment with authentic access to critical app modules, data, assets and other objects, so they are insidious. Several threat detection models like behavior based, rule-based, activity based, and impersonation based were introduced by former research scholars to find the malicious insiders from cloud environment. The major and common limitations identified from the former malicious insider threat detection models are 1) uncertain prevention from insider threats2) overload on cloud servers due to threat detection process3) suffering from false negatives in results and 4) none of the threat detection models were comprehensive. This paper proposed the Behavior Reliance Anomaly Detection (BRAD) approach, to find the malicious insiders precisely and to address the aforementioned insider threat detection limitations with cutting-edge technologies. Experimental results with BRAD prototype specified that, the proposed BRAD is proved as reliable, scalable and comprehensive than the former behavior-based anomaly detection.
Keyword: Cloud Security, Malicious Insider Threats, Behavioral Analysis, Compact Prediction Tree (CPT), Anomaly Detection Algorithm.
Scope of the Article: Cloud Computing and Networking.