A Framework for Malicious Agent Detection in Cloud Computing Environment
Ishu Gupta, Ashutosh Kumar Singh
Abstract
Data security is one of the major challenges encountered by cloud computing. The cloud data is shared among multiple entities which can be intentionally or unintentionally revealed out by any agent to the unauthorized recipient. Therefore, it has become a necessity to detect the malicious agent for protecting shared information. In this regard, we present a framework based on the probabilistic estimation that identifies the malicious agent for minimizing the likelihood of further leakage. In the proposed model, the data is distributed among multiple agents and the allocation is performed using 2-level trees. The parameters based on probability theory are computed for malicious agent identification when the data is leaked by any agent. The experimental results achieved average probability, average success rate, and detection rate up to 1, 0.98, 0.76 respectively for the various number of agents.