Learning Hidden Influences in Large-Scale Dynamical Social Networks

Interpersonal influence estimation from empirical data is a central challenge in the study of social structures and dynamics. Opinion dynamics theory is a young interdisciplinary science that studies opinion formation in social networks and has huge potential in applications such as marketing, advertising, and recommendations. The term social influence refers to the behavioral change of individuals due to the interactions with others in a social system (for example, organizations, communities, and society in general). The advent of the Internet has made a huge volume of data easily available to measure social influence across large populations. The aim of this work is to qualitatively and quantitatively infer social influence from data, from a systems and control viewpoint. First, definitions and models of opinions dynamics are introduced, and structural constraints of online social networks are considered based on the notion of sparsity. Then, the main approaches to infer a network's structure from a set of observed data are reviewed. Finally, algorithms that exploit the introduced models and structural constraints are presented, focusing on sample complexity and computational requirements.

Automated summary

Estimating interpersonal influence from empirical data is a central challenge in studying social structures and dynamics. Opinion dynamics theory, as an interdisciplinary science, has potential applications in marketing, advertising, and recommendations. The aim is to infer social influence qualitatively and quantitatively from data, with a focus on opinions dynamics models, structural constraints, and algorithms for network structure inference.