International Journal of Scientific Research and Engineering Development

The rapid expansion of digital payment platforms, including online banking, credit cards, and UPI based transactions, has transformed the modern financial ecosystem. Although these technologies provide speed and convenience, they have also increased the risk of sophisticated online financial fraud. Fraudulent activities such as identity theft, phishing attacks, transaction manipulation, and unauthorized fund transfers are becoming more complex and difficult to detect using traditional rule-based detection systems. Conventional methods rely on predefined rules and manual monitoring, which often fail to identify emerging fraud patterns in real time. This research proposes an AI-driven fraud detection system that enhances the security of online transactions using machine learning techniques. The system analyses multiple transaction attributes, including transaction amount, frequency, geographical location, device information, and user behavioural patterns. By training classification models on historical transaction datasets, the system learns to differentiate between legitimate and fraudulent transactions effectively. The proposed architecture follows a three-tier structure consisting of a presentation layer, an application processing layer, and a secure data storage layer. Data preprocessing techniques such as normalization, feature selection, and dataset balancing are applied to improve model accuracy and reliability. Experimental results show that the proposed model achieves higher detection accuracy and reduces false positive rates compared to traditional rule-based approaches. The system can identify suspicious transactions in real time and generate alerts for preventive action. The findings indicate that artificial intelligence based fraud detection significantly improves transaction security, reduces financial losses, and strengthens customer trust in digital payment systems. Furthermore, the system can adapt to evolving fraud patterns through continuous learning and future integration with advanced deep learning techniques.

Shravani Vaibhav Bhalerao,"AI-Driven Fraud detection in Online Transaction" International Journal of Scientific Research and Engineering Development, V9(3): Page(1364-1371) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.