A Bayesian Classification Model for Fraud Detection over ATM Platforms
Carolyne Milgo · IOSR Journal of Computer Engineering · 2016
The banking system relies greatly on the use of Automated Teller Machines, debit and credit cards as a vital element of its payment processing systems.The efficient functioning of payment processing systems allows transactions to be completed safely and on time thereby contributing to operational performance.Customer accounts are equally exposed to risk with respect to fraud.The major security risk identified is identity and access management of customer funds.This research sought to use classification techniques to implement a novel fraud detection solution to establish legitimacy of customer transactions.It also sought to identify the main security issues encountered with use of ATM cards and establish internal control mechanisms which can be implemented to deter possibility of card fraud.The data was obtained from a local bank and was consequently preprocessed and fed into WEKA which was used to develop the training model which was to be used for classification of incoming transactions.