A user behavior influence model of social hotspot under implicit link.

Yunpeng Xiao,Na Li,Ming Xu,Yanbing Liu
DOI: https://doi.org/10.1016/j.ins.2017.02.035
2017-01-01
Abstract:In social networks, user behavior is affected by complex dynamic factors. Here, we investigate the internal and external factors that drive users to participate in social hotspot s. By analyzing user behavior, we discover the differences between driving factors and quantify their driving strength. First, four factors that influence the users behavior are proposed, including explicit links (E), implicit links (I), personal interest (P), and a random factor (R). In particular, based on a cloud model, an implicit link creation method is designed. This method can quantify the driving strength of the implicit relation between users, and avoid the multiple attribute weighting defects in subjective and objective aspects. Next, considering the maximum likelihood estimation theory, a user behavior influence model (EIPR) of a hotspot topic is proposed to measure the causes of user behavior behind the social hotspots. Experimental results show that the model can be used to find different dynamic factors of user behavior in social hot topics. Among these external factors, the implicit link plays an significantly important role in driving user behavior.
What problem does this paper attempt to address?