Analyzing User Behavior Using Keystroke Dynamics to Protect Cloud from Malicious Insiders

Nowadays cloud computing is growing vastly as the number of companies are depending mostly on this technology due to its advantages. All the data are stored across the globe and maintained by the cloud service providers i.e., the responsibility of the data is in the hands of cloud providers. The data breach is a biggest problem in cloud as the data are shared across the globe and sensitive information of the customer is stored at some third party storage site. Security of the data is the major concern in cloud from outsiders as well as insiders. Insider attack is the most devastating threat due to the familiarity of the underlying system to the insiders. The proposed approach mitigates this threat by a host based user profiling technique where a key stroke dynamics is used for analyzing the user behavior and a retraining approach is also proposed as the imposter patterns are absent at the time of registration. Retraining boosts the overall system performance in mitigating the insider threat.