An Identification Method of Untrusted Interactive Behavior in ERP System Based on Markov Chain

Abstract

Enterprise Resource Planning (ERP) software system is widely used in enterprises as an advanced management system. In recent years, the information security problem of ERP software system has gradually attracted people’s attention. To solve the information security problem of the ERP software system, we first need to pay attention to the untrusted interactive behavior in the ERP software system. Enterprise network users generate a lot of interactive behavior in the process of using ERP system. Untrusted interactive behavior will cause huge damage to the enterprise if they are not identified. Based on this, this paper proposes a method based on Markov chain to identify untrusted interactive behavior of users in the ERP system, Firstly, a series of network user behavior characteristics are constructed based on the log records of ERP system. Then, the hidden Markov model is used to model the behavior of trusted users based on these behavior characteristics. Next, the forward algorithm is used to calculate the probability of a series of observation sequences of trusted users and untrusted users based on the hidden Markov model of trusted users. Finally, the untrusted users are identified by comparing the observation sequence probability set of trusted and untrusted users. The recognition rate of the model for trusted users is 92.64%, and the false positive rate for untrusted users is 0.76%. This result indicates that the model is effective for identifying untrusted interaction behavior.