ATM Card Fraud Detection System Using Machine Learning Techniques

Abstract

Financial fraud is an ever-growing menace with far consequences in the financial industry. ATM card fraud detection, which is a data mining problem, becomes challenging due to two major reasons. First, the profiles of normal and fraudulent behaviour change constantly and secondly, ATM card fraud data sets are highly skewed. The performance of fraud detection in ATM card transactions is greatly affected by the sampling approach on dataset, selection of variables and detection technique used. ATM card fraud causes disruptions in the digital payment and banking sector. Machine Learning is quickly emerging as the standard for mitigating risks occurring due to the use of ATM cards. This paper explores the performance of Decision Tree, Logistic Regression on largely imbalanced data sourced from European cardholders containing over 2,00,000 transactions. The work is implemented in R language using RStudio along with a GUI application developed using Shiny, which is a framework for writing rich web apps using R.