International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 03 | Mar 2025
p-ISSN: 2395-0072
www.irjet.net
Integrating Credit Card Fraud Detection with Machine Learning Algorithms Harshita Kandpal1, Talha Usmani1, Ayaan Khan1, Bakhtiyaar Khan1, and Dr. Ankita Srivastava2 1Student, Department of Computer Science & Engineering, Integral University, Lucknow INDIA 2Assistant Professor, Department of Computer Science & Engineering, Lucknow INDIA
------------------------------------------------------------------------***------------------------------------------------------------------------ABSTRACT Credit card fraud is a huge threat to financial security, making it imperative that strong fraud detection models be built. This work investigates the use of machine learning algorithms for fraud detection in transactions with high precision. We prepare the dataset for use with feature scaling methods like Standard Scaler to improve the performance of models. Different machine learning models, such as Logistic Regression, Decision Trees, Random Forest, and Support Vector Machines, are used and compared in terms of performance metrics like accuracy, precision, recall, and F1-score. The comparison indicates the advantages and disadvantages of each model, and the most suitable method for fraud detection is suggested. Experimental findings confirm that although higher accuracy is found in some models, others present better recall that is essential to reduce false negatives in fraud identification. This paper adds to efforts aimed at making financial security even better by defining the most appropriate machine learning algorithm for credit card fraud detection.
Keywords: Credit Card Fraud Detection, Machine Learning, Logistic Regression, Fraud Prevention, Model Comparison
1. INTRODUCTION Credit card fraud is a serious issue that affects numerous individuals and companies worldwide. The scammers employ different techniques, such as stolen card information, identity theft, and counterfeit transactions, to purchase goods without authorization. The greater number of individuals making electronic payments, the more sophisticated the methods become for commiting fraud, rendering the previous methods of fraud detection obsolete. To address this issue, machine learning has proven to be a robust technique for identifying suspicious transactions. Unlike other approaches, machine learning algorithms process huge volumes of transaction data, recognize patterns, and learn to identify genuine transactions and fraudulent ones. With each additional data set processed, the models keep on improving, enabling transactions to be detected at a faster and more precise rate. In this research, we investigate how machine learning, specifically logistic regression, can be used to improve credit card fraud detection. We preprocess transaction data, perform feature scaling, and measure the performance of the model using important metrics like accuracy, precision, recall, and F1-score. The aim is to create an effective fraud detection system that reduces false positives while accurately detecting fraudulent transactions. This study emphasizes the significance of sophisticated fraud detection methods in the banking industry and shows how machine learning can assist in safeguarding customers and companies against financial loss.
Figure 1- Fraud Detection System
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 752