Abstract
ObjectiveSocial media exhibit rich yet distinct temporal dynamics which cover a wide range of different scales. In order to study this complex dynamics, two fundamental questions revolve around (1) the signatures of social dynamics at different time scales, and (2) the way in which these signatures interact and form higher-level meanings.MethodIn this paper, we propose the Recursive Convolutional Bayesian Model (RCBM) to address both of these fundamental questions. The key idea behind our approach consists of constructing a deep-learning framework using specialized convolution operators that are designed to exploit the inherent heterogeneity of social dynamics. RCBM’s runtime and convergence properties are guaranteed by formal analyses.ResultsExperimental results show that the proposed method outperforms the state-of-the-art approaches both in terms of solution quality and computational efficiency. Indeed, by applying the proposed method on two social network datasets, Twitter and Yelp, we are able to identify the compositional structures that can accurately characterize the complex social dynamics from these two social media. We further show that identifying these patterns can enable new applications such as anomaly detection and improved social dynamics forecasting. Finally, our analysis offers new insights on understanding and engineering social media dynamics, with direct applications to opinion spreading and online content promotion.
Highlights
All activities in social networks evolve over time
While the formulation of Recursive Convolutional Bayesian Model (RCBM) is general enough to consider the heterogeneity of social signals, its runtime performance and solution quality are analyzed formally and confirmed experimentally
The social dynamics in Yelp are characterized by signatures representing how different groups of reviewers rate individual businesses
Summary
All activities in social networks evolve over time. Understanding the structures behind social dynamics represents a central question in social networks research, with many important applications including political campaigning [1], viral marketing [2], and disaster response [3]. While several recent works have investigated methods to identify patterns of social dynamics [4,5,6,7], in this paper, we study a new, unexplored perspective of social dynamics, namely, multi-scale compositionality. Studying multi-scale compositionality consists of identifying the compositional structures of social media dynamics, which generally covers two tasks: T1. Identification of multi-scale signatures, which consists of identifying distinct signatures across a range of time scales, as opposed to sticking with a single one; T2. Mining of compositional interactions, which requires discovering the interaction among multiple such signatures that produce higher-level meanings
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
