Evaluation of Selected Machine Learning Techniques in Feature Extraction based Fraud Detection System in Online Transactions

Abstract: The recent advances of e-commerce and e-payment systems have sparked an increment in financial fraud cases such as credit card fraud. Several classification techniques have been employed to detect credit card frauds in online transactions but their performances were affected by high cardholder’s data dimensionality. Thus, work employed Ant Colony Optimization for features extraction and evaluate its effectiveness using three selected classifiers. to detect fraud in credit cards online transactions. 3200 cardholders data (real and simulated) dataset with mix of genuine and fraudulent transactions. Ants Colony Optimization technique was used to extract features from the transactional data. Then, fraud detection system was designed with the three selected machine learning techniques (Back Propagation Neural Network, BPNN, Support Vector Machine, SVM and Naïve Bayes, NB) for classification. The results revealed that without features selection technique, NB, BPNN and SVM produced 86.4%, 88.7%, 93.6%, for accuracy respectively and while with ACO technique, the results or NB, BPNN and SVM produced 95.3%, 96.8%, and 97.6%. Keywords: Fraud Detection System, Back Propagation Neural Network, Support Vector Machine, Naïve Bayes, Ants Colony Optimization. Title: Evaluation of Selected Machine Learning Techniques in Feature Extraction based Fraud Detection System in Online Transactions Author: Ademola Sefiu A., Ismaila W. Oladimeji, Omotosho I. O., Ismaila Folasade M. International Journal of Recent Research in Mathematics Computer Science and Information Technology ISSN 2350-1022 Vol. 11, Issue 2, October 2024 - March 2025 Page No: 1-13 Paper Publications Website: www.paperpublications.org Published Date: 07-October-2024 DOI: https://doi.org/10.5281/zenodo.13897705 Paper Download Link (Source) https://www.paperpublications.org/upload/book/Evaluation%20of%20Selected%20Machine%20Learning-07102024-1.pdf