LAUSR

Information diffusion on social media has become a major approach in people's daily communication, and the value contained therein holds great interest for both academic and industrial communities. However, the process of information diffusion is affected by many factors, and the complexity of that process has not been fully explored. Most previous studies have concentrated on the strategies and driving forces in social media operations, as well as the identification of influential seed nodes, yet analyses of consumer behavior choice in the process of information diffusion are rare. Thus, This study proposes a multipoint cross-diffusion model based on MapReduce, which improves the single-point model and can better describe the product information diffusion process. On that basis, a Bayesian network model of product information diffusion was constructed to analyze the associations between factors and consumer behaviors. Moreover, the posterior probability of consumer behavior choice affected by a series of factors in the information forwarding process was considered and analyzed. This study's findings can be used to estimate the posterior probability that users will purchase, forward, or stay silent, thereby predicting the effect of product diffusion and obtaining the quantitative relationships between factors and consumer behavior.