International Research Journal of Education and Technology

In the rapidly expanding realm of e-commerce, ensuring the authenticity of products, vendors, and transactions remains a significant challenge. With the increasing sophistication of counterfeiters and fraudulent entities, traditional verification mechanisms often fall short. This project introduces a novel fusion-based approach to enhance authenticity in e-commerce platforms by integrating multiple data sources such as user behavior, transaction metadata, product reviews, and seller history using advanced machine learning and blockchain technologies. By employing a hybrid architecture that fuses supervised learning algorithms with decentralized verification protocols, the system aims to accurately distinguish between genuine and fraudulent entities. The framework leverages data fusion techniques to correlate signals across various dimensions, thus providing a more holistic and robust authenticity score. The proposed model improves trust and transparency across the supply chain while enabling consumers to make more informed decisions. Experimental results demonstrate improved detection accuracy and reduced false positives compared to existing standalone methods, thereby marking a significant advancement in secure and trustworthy e-commerce ecosystems.

Keywords: Experimental, machine, learning, architecture.