Predicting Product Adoption in Large Social Networks for Demand Estimation

Recent advances in social media that allow better access to social networks of customers have profound technical and economic implications for design and innovation research. This paper focuses on product adoption probability prediction in the context of large social networks, which paves the way for demand modeling. Traditional approaches to demand modeling in design are mainly based on customer preferences, such as utility theory or discrete choice models, which neglect the importance of explicitly modeling of customers purchase (i.e., adoption) decisions. The paper presents an integrated approach to quantify social network effects on customer choice behavior and predict product adoption probabilities through data mining from large-scale social media data.