Fraud detection on event log of bank financial credit business process using Hidden Markov Model algorithm

Abstract

Criminal cases of banks have risen by 55% in 2016. One of the reasons is the fraud in business processes cannot be detected early. Responding to that issue, this research proposes a method for detecting fraud on business processes in the bank credit application. This method uses Hidden Markov Models and activity information recorded in the event log. Hidden Markov Model that used for calculating probability possibility of fraud based on the event log. The results show that HMM method can detect fraud appropriately. The experimental results also show that the accuracy of the results is 94%.