Research on the Factors Influencing the Co-purchase Network of Products on E-commerce Platforms

As the recommender system of co-purchase network is more and more widely used, there are certain limitations in studying the economic significance of co-purchase link only based on the endogenous structure variable of network. Therefore, the exogenous variable of network word-of-mouth is added to make a more comprehensive analysis. The exponential random graph model in the social network method is used to build the model, mainly focusing on the four factors of product sales volume, product indegree, poor evaluation rate and the number of comments, to explore its influence on the formation of co-purchase links in the co-purchase network. Among them, sales volume, product indegree and comments have a positive correlation with the formation of co-purchase links, but the rate of poor reviews will significantly reduce the probability of co-purchase. The co-purchase network constructed by the exponential random graph model provides a reference for the optimization design of online e-commerce platform management network word-of-mouth and recommendation system.