Information Diffusion in Social Networks: Friendship Paradox Based Models and Statistical Inference

Abstract

Dynamic models and statistical inference for the diffusion of information in social networks is an area which has witnessed remarkable progress in the last decade due to the proliferation of social networks. Modeling and inference of diffusion of information has applications in targeted advertising and marketing, forecasting elections, predicting investor sentiment and identifying epidemic outbreaks. This chapter discusses three important aspects related to information diffusion in social networks: (i) How does observation bias named friendship paradox (a graph theoretic consequence) and monophilic contagion (influence of friends of friends) affect information diffusion dynamics. (ii) How can social networks adapt their structural connectivity depending on the state of information diffusion. (iii) How one can estimate the state of the network induced by information diffusion. The motivation for all three topics considered in this chapter stems from recent findings in network science and social sensing. Further, several directions for future research that arise from these topics are also discussed.

Controlling Segregation in Social Network Dynamics as an Edge Formation Game

“…1) The edge formation protocol proposed in this paper assumes homogeneity among users in one community, i.e., users' community information solely determines their edge formation probabilities. An interesting future direction is to incorporate heterogeneity (e.g., preferential attachment with fitness) into the network model and apply methods such as friendship paradox sampling [34], [35], [36] to assign different weights to the 2 hop connections between different pairs of users. 2) Another interesting direction is to consider ARM's parameter (e.g., the recommendation acceptance probability) depends on users' actions and incorporates a feedback law, i.e., ARM raises the acceptance probability when users become more segregated in network.…”

"What Do Your Friends Think?": Efficient Polling Methods for Networks Using Friendship Paradox

“…Our results for the case 1 and case 2 also fall under this broad theme. Apart from the applications in estimation problems, friendship paradox has also been explored in the contexts of perception biases in social networks [18], [19], [20], information diffusion and opinion formation [21], [22], [23], [24], influence maximization and stochastic seeding [25], [26], [27], node properties other than the degrees [28], [29], [30] and directed social networks [18], [28], [31].…”

Maximum Likelihood Estimation of Power-law Degree Distributions via Friendship Paradox based Sampling

“…intending to vote for a certain political party) on an undirected social network. Apart from these, [23][24][25][26][27][28][29][30][31][32][33][34][35] also explore further effects and generalizations of friendship paradox.…”