Credit Card Fraud Detection Using Autoencoder Model in Unbalanced Datasets
Journal of Advances in Mathematics and Computer Science,
Page 1-16
DOI:
10.9734/jamcs/2019/v33i530192
Abstract
Fraudulent credit card transaction is still one of problems that face the companies and banks sectors; it causes them to lose billions of dollars every year. The design of efficient algorithm is one of the most important challenges in this area. This paper aims to propose an efficient approach that automatic detects fraud credit card related to insurance companies using deep learning algorithm called Autoencoders. The effectiveness of the proposed method has been proved in identifying fraud in actual data from transactions made by credit cards in September 2013 by European cardholders. In addition, a solution for data unbalancing is provided in this paper, which affects most current algorithms. The suggested solution relies on training for the autoencoder for the reconstruction normal data. Anomalies are detected by defining a reconstruction error threshold and considering the cases with a superior threshold as anomalies. The algorithm's performance was able to detected fraudulent transactions between 64% at the threshold = 5, 79% at the threshold = 3 and 91% at threshold= 0.7, it is better in performance compare with logistic regression 57% in unbalanced dataset.
Keywords:
- Autoencoder
- fraudulent credit card
- machine learning
- logistic regression.
How to Cite
References
The Nilson report; 2016.
Available:http://www.nilsonreport.com
LexisNexis. True cost of fraud 2016 study.
Available:https://risk.lexisnexis.com/insights-resources/research/lexisnexis-2016-true-cost-of-fraud
Argaam report; 2018.
Available:https://www.argaam.com/en/article/articledetail/id/570536
Yeh IC, Lien C. The comparisons of data mining techniques for the predictive accuracy of probability of default of credit card clients. Expert Systems with Applications. 2009;36(2):2473-2480.
West J, Bhattacharya M, Islam R. Intelligent financial fraud detection practices: An investigation. In International Conference on Security and Privacy in Communication Systems. Cham: Springer. 2014; 186-203.
Bhatla PT, Vikram P, Amit D. Understanding credit card frauds. Cards Business Review. 2003;1:6.
Bose I, Wang J. Data mining for detection of financial statement fraud in Chinese Companies. In International joint Conference on e-Commerce, e-Administration, e-Society, and e-Education. International Business Academics Consortium (IBAC) and Knowledge Association of Taiwan (KAT). Taiwan; 2007.
Kirkos E, Spathis C, Manolopoulos Y. Data mining techniques for the detection of fraudulent financial statements. Expert Systems with Applications. 2007;32(4):995-1003.
Ravisankar P, Ravi V, Rao GR, Bose I. Detection of financial statement fraud and feature selection using data mining techniques. Decision Support Systems. 2011;50(2):491-500.
Bhattacharyya S, Jha S, Tharakunnel K. Data mining for credit card fraud: A comparative study. Decision Support Systems. 2011;50(3):602-613.
Pinquet J, Ayuso M, Guillen M. Selection bias and auditing policies for insurance claims. Journal of Risk and Insurance. 2007;74:425-40.
Ngai EWT, Hu Y, Wong YH, Chen Y, Sun X. The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literature. Decision Support Systems. 2011;50(3):559-569.
Chawla NV, Bowyer KW, Hall LO, Philip KPW. SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research. 2002;16:321-357.
A Dal Pozzolo. Adaptive machine learning for credit card fraud detection; 2015.
Zhao X, Zhang J, Qin X. LOMA: A local outlier mining algorithm based on attribute relevance analysis. Expert Systems with Applications. 2017;84:272-280.
Bahnsen AC, Aouada D, Stojanovic A, Ottersten B. Feature engineering strategies for credit card fraud detection. Expert Systems with Applications. 2016;51:134-142.
Van Vlasselaer V, Bravo C, Caelen O, Eliassi-Rad T, Akoglu L, Snoeck M, Baesens B. APATE: A novel approach for automated credit card transaction fraud detection using network-based extensions. Decision Support Systems. 2015;75:38-48.
Graves A. Supervised sequence labelling. In Supervised sequence labelling with recurrent neural network. Berlin, Heidelberg: Springer. 2012;5-13.
Wang Y, Xu W. Leveraging deep learning with LDA-based text analytics to detect automobile insurance fraud. Decision Support Systems. 2018;105:87-95.
Patidar R, Sharma L. Credit card fraud detection using neural network. International Journal of Soft Computing and Engineering (IJSCE). 2011;10:32-38.
Şahin YG, Duman E. Detecting credit card fraud by decision trees and support vector machines. Proceeding of International Multi-conference of Engineering and Computer Statistics; 2011.
Alzubi J, Nayyar A, Kumar A. Machine learning from theory to algorithms: An overview. In Journal of Physics: Conference Series. IOP Publishing. 2018;1142(1): 012012.
X. Niu, Wang L, Yang X. A comparison study of credit card fraud detection: Supervised versus unsupervised. arXiv preprint arXiv:1904.10604; 2019.
Yu S, Jenssen R, Principe JC. Understanding convolutional neural network training with information theory. arXiv preprint arXiv:1804.06537; 2018.
Download the credit card fraud dataset.
Available:https://www.kaggle.com/mlg-ulb/creditcardfraud/data
Tasche D. A plug-in approach to maximising precision at the top and recall at the top. arXiv preprint arXiv:1804.03077; 2018.
Mekterović I, Brkić L, Baranović M. A systematic review of data mining approaches to credit card fraud detection. WSEAS Transactions on Business and Economics. 2018;15:437.
-
Abstract View: 3804 times
PDF Download: 1947 times