Improving Credit Card Fraud Detection in Imbalanced Datasets: A Comparative Study of Machine Learning Algorithms
Ruozhang Liu
2024
Abstract
This research focuses on tackling the challenge of identifying credit card fraud within highly imbalanced datasets, where the proportion of fraudulent transactions is significantly smaller compared to the overall number of transactions. Using a dataset from the Kaggle, this study applied various preprocessing techniques, including normalization, data cleaning and undersampling and so on to balance the data. This paper aims to evaluate several machine learning algorithms—Logistic Regression, K-Nearest Neighbors, Support Vector Machines (SVM), and Decision Trees—based on different metrics. Logistic Regression and SVM showed the best performance, balancing precision and recall effectively. Despite improvements, the trade-off between precision and recall remains a challenge, indicating the need for more advanced methods like ensemble learning and deep learning. The findings emphasize the importance of sophisticated machine learning techniques in improving the accuracy and reliability of credit card fraud detection systems, ultimately protecting financial institutions and customers from significant financial losses.
DownloadPaper Citation
in Harvard Style
Liu R. (2024). Improving Credit Card Fraud Detection in Imbalanced Datasets: A Comparative Study of Machine Learning Algorithms. In Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI; ISBN 978-989-758-726-9, SciTePress, pages 393-397. DOI: 10.5220/0013263900004568
in Bibtex Style
@conference{ecai24,
author={Ruozhang Liu},
title={Improving Credit Card Fraud Detection in Imbalanced Datasets: A Comparative Study of Machine Learning Algorithms},
booktitle={Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI},
year={2024},
pages={393-397},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013263900004568},
isbn={978-989-758-726-9},
}
in EndNote Style
TY - CONF
JO - Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI
TI - Improving Credit Card Fraud Detection in Imbalanced Datasets: A Comparative Study of Machine Learning Algorithms
SN - 978-989-758-726-9
AU - Liu R.
PY - 2024
SP - 393
EP - 397
DO - 10.5220/0013263900004568
PB - SciTePress