Research on E-Commerce Social Relationship Modeling and Marketing Path Optimization Method Based on Graph Neural Network
Q. L. Li
Hunan University of Arts and Science
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
With the deep integration of social media and e-commerce, social relations have become a core driving force affecting consumer purchase decisions within the textile and apparel market, where digital word-of-mouth shapes users’ perceptions of fabric quality, garment design, functional performance, and fashion trends. This social influence directly steers the demand for innovative textile products, including smart textiles, electromagnetic functional fabrics, wearable sensing garments, and other apparel products with technical application value, as consumer interactions increasingly affect the market acceptance of specific garment designs and material applications. Traditional e-commerce marketing methods encounter bottlenecks such as sparse data, weak social relevance, and inefficient precision marketing. To solve these problems, this study proposes a comprehensive research framework for e-commerce social relationship modeling and marketing path optimization based on a graph neural network (GNN). First, a multidimensional e-commerce social relationship graph (ESRG) is constructed by integrating explicit social connections, implicit interactive behavior, and multimodal attribute information. Second, a transitive enhanced graph attention network (TE-GAT) is designed to accurately capture the structural characteristics of social networks, transitive relationship logic, and the dynamic evolution of user preferences. Finally, based on the user preferences learned from the model, three optimized marketing paths are formed: trust-driven precise recommendation, social group-purchase marketing, and opinion-leader-oriented social fission. The proposed method provides a data-driven reference for improving marketing communication and user conversion efficiency in textile and apparel e-commerce scenarios involving both conventional fashion products and emerging electromagnetic functional textile applications.
逐年被引趋势
暂无年度引用数据
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
经济 / 管理Digital Marketing and Social Media
Recommender Systems and Techniques · Complex Network Analysis Techniques