E-Commerce Detectives: A Scam Website Detection Application Using Machine Learning

• Published in: 2024 6th IEEE Symposium on Computers and Informatics, ISCI 2024
• Main Author: Azzuri P.N.; Suffian Sulaiman M.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85204982787&doi=10.1109%2fISCI62787.2024.10667704&partnerID=40&md5=9ecef21bb17baaeededf71c5f085172a

Many people have chosen to use online-based shopping, which is easy, fast and more accessible to different types of products and services. Some of the popular e-commerce websites nowadays are Shopee, Lazada, eBay, Amazon and Alibaba. These websites serve as a good platform for people to find different kinds of products and services around the world. Most users are younger, ranging in age from 13 to 25 years old. However, some older people love to shop online. Most of them lack exposure to and education on how to tell the difference between a scam and a legitimate website. It is also one of humans' weaknesses to detect and identify a scam website if it looks the same as a legitimate website. Furthermore, there are numerous reports of scam attacks that result in significant financial losses each year. Scammers may target not just an individual but also a company. At times, the situation can worsen, potentially posing a threat to our families and lives. This study has proposed a web-based scam detector as a solution to verify the legitimacy of an e-commerce website. This prototype uses the Random Forest model as a classifier to detect scams. This is due to the good performance and high accuracy of detection compared to the other algorithms such as Decision Tree, Naïve Bayes and Support Vector Machine. The Random Forest model achieved 93% accuracy of detection and it is good enough to deploy on the real prototype. Additionally, this website offers guidance on identifying scam websites and the tactics they typically employ to lure victims. © 2024 IEEE.