Home / Current Issue / Paper 1713053
A Machine-Learning Framework: Integrating MLP And SMOTE For Deceitful Payment Classifications
Subject area: Science,Engineering and Technology · Area of research: Machine Learning
Abstract
In this study, Multiple Layer Perceptron?s (MLP) integrating Synthetic Minority Oversampling Technique (SMOTE) was offer as potential solution to the despicable problem of payment fraud. This model was design and realized on Google Colab platform embedding GPU, where Tensorflow was fit for deep learning (DL), and Scikit learn for machine learning (ML) model respectively. Python was the modeling language. Firstly, baseline experiment was conduct on orthodox ML models of Random Forest (RF), Logistics Regression (LR), Isolation Forest (iforest) with MLP for feature selection and engineering. While, probing kaggle dataset. This dataset is highly imbalance, a bait to the ML challenges suffering in the baseline trial. However, resampling method of SMOTE was applied in the second trial to balance the dataset, address over-fitting with under-fitting problem and improved on the utilized model performances. In testing for these models? efficacy, confusion matrix and performance evaluation metric were explored. This revealed the outcome of the balancing model trial, where the proposed MLP+SMOTE exercises superclass performance against other models. Presenting accuracy score of 0.95%, Error Rate of 0.05%. Recall of 0.98%, least False Positive Rate of 0.07%, True Negative Rate of 0.93%, Precision of 0.93%, Prevalence of 0.46%, Null Error Rate of 100%, Cohen kappa of 0.42%, F1-Score of 0.95% and Matthews Correlation Coefficient of 0.91% respectively. This result validates the developed model as amazing in performance, when compared with benchmark studies; and it is promising in the classification of deceitful payment transaction.
Keywords
Payment card, Credit card fraud, Confusion Matrix, Performance Evaluation Metrics
References
[1] Aghware FO, Yoro R E, Ejeh P O, Odiakaose C C, Emordi F U, Ojugo, A A. DeLClustE: Protecting Users from Credit-Card Fraud Transaction via the Deep-Learning Cluster Ensemble, (IJACSA) International Journal of Advanced Computer Science and Applications, 2023, Vol. 14 No.6, pp. 94-100. www.ijacsa.thesai.org
[2] Akinola K E, Aina D A, Oyede O, Braimoah J A (2023). Credit Card Fraud Detection Using Logistics Regression and Isolation Forest Algorithm, UNIZIK Journal of Engineering and Applied Sciences, 2023, Vol. 2 No.1, pp. 187-195. https://journals.unizik.edu.ng/index.php/ujeas
[3] Alraddadi, B. A Survey and a Credit Card Fraud Detection and Prevention Model using the Decision Tree Algorithm, Engineering, Technology & Applied Science Research,2023, Vol. 13 No.4, pp. 11505- 11510. http://www.etasr.com/
[4] Al-Smadi, B. Credit card security system and fraud detection algorithm, Ph.D. dissertation, College of Engineering and Science, Louisiana Tech University, USA, 2021. https://digitalcommons.latech. edu/cgi/viewcontent.cgi?article=1947&context =dissertations.
[5] Aslam A, Hussain A. A performance Analysis of Machine Learning Techniques for Credit Card Fraud Detection”, Journal on Artificial Intelligence. 2024. https://
[6] Bokhare A, Bhagat A, Bhalodia, R. Multilayer Perceptron’s for heart Failure Detection using SMOTE Techniques, Springer Nature. 2023. Retrieved from: Multi- layer Perceptron for Heart Failure Detection Using SMOTE Technique | SN Computer Science (springer.com)
[7] Brownlee J. 14 Different Types of Learning in Machine Learning, Blog, 2019. https://machinelearningmastery.com/types-of- learning-in-machine-learning/
[8] Cheon M J, Lee D H, Joo S H, Lee O. Deep learning based hybrid approach of detecting fraudulent transactions Journal of theoretical and applied information technology 31st august 2021. Vol. 99 No.16, pp. 4044-4054. Http://www.jatit.org/
[9] Cicekli I. Classification Model Evaluation and Selection and ensemble Methods, Data Mining. 2022. https://www.lec07_Classification_ModelEvalu ation_Ensemble.pdf
[10] Comotto F. Evaluation metric: leave your comfort zone ant try MCC and brier scope.2022.
[11] Cortes, D. An Introduction to Isolation Forest. [Blog].2019. https://cran.r- project.org/web/packages/isotree/vignettes/An _introduction_to_isolation_forest.html#:testiso lation%20%forest20is%20an%20unsupervised %2ffeatures%2fcolumn%20at%20random
[12] DataCam. Hyperparameter turning of Isolation forest. Nd. https://www.campus.datacamp.com/courses/an omaly-detection-in-python-/isolation-forests- with-pyod?ex=9
[13] Devi R R, & Parthibranjanray. A., Credit Card Fraud Detection using AI/ML/CNN, learning and data augmentation techniques, Indatu Journal of Management and Accounting,2023, Vol.1 No.1. https://heca- analitika.com/ijm
[14] Pavan P. Random Forest Hyperparameter Tuning in Python: Complete Guide with Examples. 2024.https:// www.upgrade.com/blog/random- forest--hyperparameter-turning/
[15] Prasad P Y, Chowdary A S, Bavitha C, Mounisha, E, & Reethika C. A comparison study of fraud detection in usage of credit cards using machine learning, In Proceedings of the 7th International Conference on Trends in Electronics and Informatics (ICOEI), Tirunelveli, India, 11–13, April 2023; pp. 1204–1209.
[16] Salaudeen L G, Gabi, D, Muhammad, G, & Suru, H. U. A boosted deep convnet embedding long short-term memory with synthetic minority oversampling techniques as foiling model for payment card fraud, Journal of International Financial Trends. 2025 Vol.1, No.1, pp.61 -89. https://
[17] Salaudeen L G, Gabi, D, Muhammad, G, & Suru, H. U. Light gradient boosting machine (LGBM) for credit card fraud detection in financial institution, Direct Res. J. Eng. Inform. Tech. 2024a, Vol. 12 No.1, pp. 19-34. https://
[18] Salaudeen, L G, Gabi, D, Muhammad G, & Suru, H U. Deep convolutional neural network (DCNN) based synthetic minority oversampling techniques: a forfending model for fraudulent credit card transactions in financial institution, Journal of Nigerian Society of Physical Sciences (NSPS). 2024b. https://
[19] Sharma, A., Free GPUs for Everyone! Get Started with Google Colab for Machine Learning and Deep Learning. 2020. Retrieved from: https://www.analyticsvidhya.com/blog/2020/0 3/google-colab-machine-learning-deep- learning
[20] Singh S, Ninje, H, Ajinkya F, Neware R. Credit card fraud detection using a hybrid machine learning algorithm. 2024. https:/// v1
How to cite this paper
@article{1713053,
author = {Lateef G. Salaudeen, Babatola Moses Omilodi, Oduntan Esther Odunnayo, Michael O. Asanbe, Omobolanle Esther Akinjisola; Temitope Mosunmola Olatunji; Emmanuel Abiodun},
title = {A Machine-Learning Framework: Integrating MLP And SMOTE For Deceitful Payment Classifications},
journal = {Iconic Research And Engineering Journals},
year = {2025},
volume = {9},
number = {6},
pages = {1916-1934},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1713053.pdf},
abstract = {In this study, Multiple Layer Perceptron?s (MLP) integrating Synthetic Minority Oversampling Technique (SMOTE) was offer as potential solution to the despicable problem of payment fraud. This model was design and realized on Google Colab platform embedding GPU, where Tensorflow was fit for deep learning (DL), and Scikit learn for machine learning (ML) model respectively. Python was the modeling language. Firstly, baseline experiment was conduct on orthodox ML models of Random Forest (RF), Logistics Regression (LR), Isolation Forest (iforest) with MLP for feature selection and engineering. While, probing kaggle dataset. This dataset is highly imbalance, a bait to the ML challenges suffering in the baseline trial. However, resampling method of SMOTE was applied in the second trial to balance the dataset, address over-fitting with under-fitting problem and improved on the utilized model performances. In testing for these models? efficacy, confusion matrix and performance evaluation metric were explored. This revealed the outcome of the balancing model trial, where the proposed MLP+SMOTE exercises superclass performance against other models. Presenting accuracy score of 0.95%, Error Rate of 0.05%. Recall of 0.98%, least False Positive Rate of 0.07%, True Negative Rate of 0.93%, Precision of 0.93%, Prevalence of 0.46%, Null Error Rate of 100%, Cohen kappa of 0.42%, F1-Score of 0.95% and Matthews Correlation Coefficient of 0.91% respectively. This result validates the developed model as amazing in performance, when compared with benchmark studies; and it is promising in the classification of deceitful payment transaction.},
keywords = {Payment card, Credit card fraud, Confusion Matrix, Performance Evaluation Metrics},
month = {December},
doi = {https://doi.org/10.64388/IREV9I6-1713053}
}