Home / Current Issue / Paper 1720044
A Machine Learning Framework for Financial Fraud Detection Using Credit Card Transaction Data
Subject area: Science,Engineering and Technology · Area of research: Machine Learning
Abstract
Financial fraud has become one of the most significant challenges confronting modern electronic payment systems, resulting in substantial financial losses for individuals and financial institutions worldwide. The increasing volume of digital financial transactions necessitates intelligent fraud detection systems capable of accurately distinguishing fraudulent transactions from legitimate ones. This study proposes a machine learning framework for financial fraud detection using a publicly available credit card transaction dataset obtained from Kaggle. The research focuses on analyzing the characteristics of financial transaction data and developing a robust framework for fraud detection through systematic data preprocessing and feature engineering. The methodology involved duplicate removal, feature scaling using StandardScaler, creation of a transaction-hour feature and class balancing using the Synthetic Minority Over-sampling Technique (SMOTE) to address the severe class imbalance inherent in fraud datasets. The processed data were integrated into a structured machine learning framework designed to support fraud detection and future model deployment. The proposed framework provides a comprehensive workflow encompassing data acquisition, preprocessing, feature engineering, model development and evaluation. The developed architecture enhances data quality, improves model readiness, and establishes a reliable foundation for implementing intelligent fraud detection systems in financial environments. The study contributes a scalable and practical machine learning framework that can be adapted for real-world financial transaction monitoring and future intelligent fraud prevention applications.
Keywords
Financial Fraud Detection, Machine Learning, Credit Card Transactions, Data Preprocessing, SMOTE
How to cite this paper
@article{1720044,
author = {Yahuza Nafiu, Prof. Ahmad Baita Garko, Dr. Abubakar Atiku Muslim},
title = {A Machine Learning Framework for Financial Fraud Detection Using Credit Card Transaction Data},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {10},
number = {1},
pages = {2110-2124},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1720044.pdf},
abstract = {Financial fraud has become one of the most significant challenges confronting modern electronic payment systems, resulting in substantial financial losses for individuals and financial institutions worldwide. The increasing volume of digital financial transactions necessitates intelligent fraud detection systems capable of accurately distinguishing fraudulent transactions from legitimate ones. This study proposes a machine learning framework for financial fraud detection using a publicly available credit card transaction dataset obtained from Kaggle. The research focuses on analyzing the characteristics of financial transaction data and developing a robust framework for fraud detection through systematic data preprocessing and feature engineering. The methodology involved duplicate removal, feature scaling using StandardScaler, creation of a transaction-hour feature and class balancing using the Synthetic Minority Over-sampling Technique (SMOTE) to address the severe class imbalance inherent in fraud datasets. The processed data were integrated into a structured machine learning framework designed to support fraud detection and future model deployment. The proposed framework provides a comprehensive workflow encompassing data acquisition, preprocessing, feature engineering, model development and evaluation. The developed architecture enhances data quality, improves model readiness, and establishes a reliable foundation for implementing intelligent fraud detection systems in financial environments. The study contributes a scalable and practical machine learning framework that can be adapted for real-world financial transaction monitoring and future intelligent fraud prevention applications.},
keywords = {Financial Fraud Detection, Machine Learning, Credit Card Transactions, Data Preprocessing, SMOTE},
month = {July},
}