Review of Fraud Detection Methods and Development of a Data Mining Technique for Real-Time Financial Fraud Detection.
No Thumbnail Available
Date
2022
Authors
Ogude, Ufuoma
Nwohiri, Anthony
Ugbaja, Geraldine
Journal Title
Journal ISSN
Volume Title
Publisher
Bayero University Kano
Abstract
Fraud that involves cell phones, insurance claims, tax return claims, credit card transactions, government procurement etc. represent significant problems for governments and businesses. Technology advances have brought along new opportunities and security challenges. Due to the dramatic increase in fraud which has cost businesses billions of dollars each year, several modern fraud detection techniques are continually evolving to meet the unprecedented challenge. Data mining (DM) is the most recognized and effective technology that has been deployed for fraud detection. This study looks at the concept of DM and current techniques used in detecting fraud and reviews DM methods used to detect fraudulent payment transactions. It explores some of the most effective DM techniques for detecting different types of fraud, categorizing them based on supervised and unsupervised methods. An efficient model that can differentiate fraudulent transactions from genuine transactions based on a given dataset is proposed.
Description
Keywords
Fraud Detection; Data Mining Techniques; Financial Fraud; Supervised Learning; K-Means;
Citation
Ogude, U.C., Nwohiri, A.M., Ugbaja, G.U. (2022). Review of Fraud Detection Methods and Development of a Data Mining Technique for Real-Time Financial Fraud Detection. Bayero Journal of Engineering and Technology (BJET), 17 (1), 1-11