• Avinash G M
• Neha M
• Yashaswini C M
• Arya Sudhakar Badiger
• Shubhra V Reddy

Ensuring secure transactions on e-commerce platforms has long been a critical concern, particularly with the growing sophistication of fraudulent activities. Traditional fraud detection methods primarily rely on historical order data, which often prove inadequate due to the inherent anonymity and dynamic nature of user behavior in online transactions. Many existing approaches fall short by overlooking real-time, multi-perspective behavioral analysis, resulting in reduced detection accuracy. To address these limitations, this paper presents a novel fraud detection approach that combines machine learning with process mining techniques for real-time monitoring of user activities. Initially, we construct a process model tailored to B2C e-commerce systems that incorporates behavioral tracking. Next, we introduce a method for abnormal behavior analysis, capable of extracting significant features from event logs. These features are subsequently input into a Support Vector Machine (SVM) classifier, enabling effective detection of fraudulent transactions. Experimental results validate the proposed approach’s ability to accurately capture and identify dynamic fraudulent behaviors, thereby enhancing the overall security of e-commerce platforms

E-Commerce fraud detection, support vector machine (SVM), Real-time monitoring.

"Expansion of a Comprehensive Fraud Detection System Using Multi-Dimensional Analysis", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.e589-e595, May-2025, Available :http://www.jetir.org/papers/JETIR2505539.pdf

"Expansion of a Comprehensive Fraud Detection System Using Multi-Dimensional Analysis", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. ppe589-e595, May-2025, Available at : http://www.jetir.org/papers/JETIR2505539.pdf