ONLINE TRANSACTION FRAUD DETECTION USING BACKLOGGING ON ECOMMERCE WEBSITE

Computer Science Engineering

Online transactions are becoming more common and convenient, but they also pose a risk of fraud and cyber-crime. To prevent and detect fraudulent transactions on e-commerce websites, a system using backlogging is proposed. Backlogging is a technique that blocks the application of a transaction until it is verified by the user or the bank. The system uses a behaviour and location analysis (BLA) to compare the current transaction with the user’s previous patterns and preferences. If the BLA detects any anomaly or inconsistency, the system asks for a re-verification from the user or the bank. The system also uses machine learning methods to identify and classify the types of frauds and the fraudsters. The system aims to reduce the false positive and false negative rates of fraud detection, and to enhance the security and trust of online transactions. The system is implemented using Python and tested on a simulated e-commerce website. The results show that the system can effectively detect and prevent various types of frauds, such as identity theft, card cloning, phishing, and spoofing. The system also provides a user-friendly interface and a feedback mechanism for the users and the banks. The system can be integrated with existing e-commerce platforms and can be customized according to the needs and preferences of the users and the banks. If any sort of surprising pattern is detected by the FDS then it asks for a re- verification. The algorithm used in the system then analyses all previous information of that card holder and recognizes any unusual pattern in the payment procedure.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.