Analysis of E-commerce user social network influence based on enhanced graph attention network algorithm

This research employs an advanced Graph Attention Network (GAT) model to assess user influence in e-commerce social networks. As social networks become increasingly embedded in online shopping environments, it is essential for businesses to comprehend how user relationships, behaviors, and interactions shape purchasing behavior in order to refine marketing tactics. Conventional influence analysis techniques are inadequate for handling the intricate and varied nature of these networks, which include diverse user engagements and transaction records. This research proposes a modified GAT model tailored to e-commerce environments, which incorporates advanced attention mechanisms to better capture the significance of various user interactions, product categories, and demographic attributes. The study first constructs a graph-based representation of the e-commerce social network by integrating transactional, demographic, and social connection data. It then applies the enhanced GAT model to quantify user influence, emphasizing the importance of contextual interactions and varying user behaviors. Experimental results, validated using real-world e-commerce data, show that the modified GAT model outperforms traditional influence metrics in terms of accuracy, offering businesses actionable insights into identifying key influencers within their networks. Key contributions of the research include the development of a comprehensive graph-based framework for influence analysis, the introduction of domain-specific modifications to GAT, and the empirical validation of the proposed model’s effectiveness. These developments allow for more precise detection of key influencers, supporting e-commerce platforms in enhancing marketing efficiency, strengthening customer interaction, and boosting sales outcomes. By introducing a new method of influence analysis specifically designed for e-commerce social networks, this study addresses important gaps in existing research and provides a foundation for future investigations in the field.