Credit Card Fraud Detection - A Machine Learning Perspective

Shahana Fathima, Leena C Sekhar, Jaseena K U

Abstract: The increasing prevalence of credit card fraud in today's digital economy poses a significant challenge to financial institutions and consumers alike. To combat this threat, there is a growing need for robust and efficient fraud detection systems. This paper presents a comprehensive machine learning approach for credit card fraud detection, leveraging advanced techniques and models to enhance the accuracy and reliability of fraud detection mechanisms. Our methodology encompasses data pre - processing, feature engineering, and model selection to construct a highly effective fraud detection pipeline. We explore various machine learning algorithms, including K Nearest Neighbour, Support Vector Machine, Random Forest, Decision Trees and Artificial Neural Networks, to build a predictive model that can distinguish between legitimate and fraudulent credit card transactions. The dataset used for training and evaluation is sourced from historical credit card transaction records, encompassing a wide range of transaction attributes and labels for fraudulent and non - fraudulent activities. We apply rigorous performance metrics, such as precision, recall, F1 - score, to assess the models' efficacy. The proposed model achieves high accuracy rates while minimizing false positives, thus enhancing the overall security of credit card transactions.

Keywords: Credit Card Fraud, Fraud Detection, Feature Selection, RandomForest, K - Nearest Neighbours, Artificial Neural Network

How to Cite?: Shahana Fathima, Leena C Sekhar, Jaseena K U, "Credit Card Fraud Detection - A Machine Learning Perspective", Volume 12 Issue 11, November 2023, International Journal of Science and Research (IJSR), Pages: 2002-2009, https://www.ijsr.net/getabstract.php?paperid=SR231128212352, DOI: https://dx.doi.org/10.21275/SR231128212352

Download Citation: APA | MLA | BibTeX | EndNote | RefMan