Analysis of Supervised Machine Learning Algorithms in the Context of Fraud Detection

• Published in: ECS transactions Vol. 107; no. 1; pp. 7189 - 7200
• Main Authors: Verma, Pradeep, Tyagi, Poornima
• Format: Journal Article
• Language: English
• Published: The Electrochemical Society, Inc 24.04.2022
• ISSN: 1938-5862; 1938-6737
• DOI: 10.1149/10701.7189ecst

In today’s era, where ‘time’ is considered as ‘money,’ people are completely depending on e-commerce and online banking for their routine purchases, shopping, and financial transactions. This increasing dependency on e-commerce are increasing fraud in online transactions, and credit card fraud is one example. Such malicious and unethical practices may cause identity theft and monitory loss to the people across the world. In this research paper, our effort is to identify the best Supervised Machine Learning algorithm that helps in classifying fraudulent and non-fraudulent transactions under credit card fraud on an imbalanced dataset. To conduct this research and compare the results, we have used five different Supervised Machine Learning Classification techniques. On implementing these machine learning techniques, it has been observed that both Supervised Vector Classifier and Logistic Regression Classifier perform better for detecting credit card fraud in an imbalanced dataset.