Cyber Criminal Activity Analysis Models using Markov Chain for Digital Forensics

Recognizing links between offender patterns is one of the most crucial skills of an investigator. Early recognition of similar patterns can lead to focusing resources, improving clearance rates, and ultimately saving lives in terms of digital forensics. In this paper we propose a forensics methodology using Markov chain during a given time interval for tracking and predicting the degree of criminal activity as it evolves over time. In other words, we describe intrusion scenario, and classify profiling of user's behavior by prior probability based Markov chain. Also, we apply the noise page elimination algorithm (NPEA) to reduce an error of probability prediction. Finally, we have experiment our model on dataset and have analysis their accuracy by Monte Carlo simulation.