Predictive Analytics and Machine Learning-Based Models for E-Commerce Fraud Prevention
Journal of Advances and Scholarly Researches in Allied Education · 01.06.2026 · 23 (3) · 267-286
DOI: 10.29070/sjdaxv21 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Advances and Scholarly Researches in Allied Education, online erschienen am 01.06.2026. Verfasst von Sachin Bagoria und 1 weiteren Autor:innen.
Abstract (Originalsprache)
The e-commerce market has grown but so have online fraud and criminality. Online marketplaces can be very complex and customers are becoming more adept at new methods of fraud, which make traditional fraud detection methods ineffective. The aim of this study is to design a machine learning and predictive analysis system for e-commerce fraud prevention. It takes into account the structure of the URL, the content of the HTML, the technology profiles, SSL certificates, HTTP headers and external reputation indicators. 2,031 ecommerce sites were used for model creation and evaluation, with 739 of them being fraudulent and 1,292 authentic. XGBoost, Odd Forest, Support Vector Machine, Logistic Regression, k-Nearest Neighbour, AdaBoost, & Naïve Bayes were used to extract and evaluate 50 features. Experimentally, XGBoost outperformed the baseline with all characteristics at 0.9688 F1-Score & 97.78 accuracy rate while the baseline is 0.9653 & 97.49%. The comparative investigation revealed the superiority of the proposed system over existing fraud detection systems. The results suggest that machine learning-based predictive analytics could be a scalable and powerful tool to protect online transactions and detect fraudulent ecommerce sites.
Zitieren
APA
Bagoria, S., & Kumari, K. (2026). Predictive Analytics and Machine Learning-Based Models for E-Commerce Fraud Prevention. Journal of Advances and Scholarly Researches in Allied Education, 23(3), 267-286. https://doi.org/10.29070/sjdaxv21
Vancouver
Bagoria S, Kumari K. Predictive Analytics and Machine Learning-Based Models for E-Commerce Fraud Prevention. Journal of Advances and Scholarly Researches in Allied Education. 2026;23(3):267-286. doi:10.29070/sjdaxv21
BibTeX
@article{Bagoria2026,
author = {Bagoria, Sachin and Kumari, Kavita},
title = {Predictive Analytics and Machine Learning-Based Models for E-Commerce Fraud Prevention},
journal = {Journal of Advances and Scholarly Researches in Allied Education},
year = {2026},
volume = {23},
number = {3},
pages = {267--286},
doi = {10.29070/sjdaxv21}
}
EUVIMED