Development and implementation of an innovative e-commerce recommendation system used in the B2B energy sector

Temporarily accepted

The advent of the digital age has transformed e-commerce platforms into key tools for the industry, but traditional recommendation systems often fail in the specialized context of the power industry. These systems typically struggle with unique industry challenges, such as infrequent and risky transactions, prolonged decision-making processes, and sparse data. This research develops a new recommendation engine tailored to these specific conditions, such as low-frequency, long-cycle Business-to-Business (B2B) transactions. The approach includes algorithmic improvements to better process and interpret the limited available data and data pre-processing techniques designed to enrich the sparse data sets specific to this industry. This research also introduces a methodological innovation that integrates multi-dimensional data by combining e-commerce user activities, product specificity, and non-tender-related essential information. The proposed engine leverages advanced machine learning techniques to provide more accurate and relevant recommendations. The results demonstrate a clear improvement over traditional models, offering a more robust and effective tool for facilitating B2B transactions in the power industry. This research not only addresses the unique challenges facing this sector, but also provides a model for adapting recommendation systems to other industries with similar B2B characteristics.