Modeling opinion dynamics in social networks

Abstract

Our opinions and judgments are increasingly shaped by what we read on social media -whether they be tweets and posts in social networks, blog posts, or review boards. These opinions could be about topics such as consumer products, politics, life style, or celebrities. Understanding how users in a network update opinions based on their neighbor's opinions, as well as what global opinion structure is implied when users iteratively update opinions, is important in the context of viral marketing and information dissemination, as well as targeting messages to users in the network.In this paper, we consider the problem of modeling how users update opinions based on their neighbors' opinions. We perform a set of online user studies based on the celebrated conformity experiments of Asch [1]. Our experiments are carefully crafted to derive quantitative insights into developing a model for opinion updates (as opposed to deriving psychological insights). We show that existing and widely studied theoretical models do not explain the entire gamut of experimental observations we make. This leads us to posit a new, nuanced model that we term the BiasedVoterModel. We present preliminary theoretical and simulation results on the convergence and structure of opinions in the entire network when users iteratively update their respective opinions according to the BiasedVoterModel. We show that consensus and polarization of opinions arise naturally in this model under easy to interpret initial conditions on the network.

Tolerance-based interaction: a new model targeting opinion formation and diffusion in social networks

“…Generally speaking, previous models use fixed thresholds (Javarone & Squartini, 2014;Biswas et al, 2011;Li et al, 2012;Das, Gollapudi & Munagala, 2014;Li et al, 2013) or thresholds extracted from real-world examples (Galuba et al, 2010;Saito et al, 2011). However, there are a few models which use dynamic thresholds (Fang, Zhang & Thalmann, 2013;Deng, Liu & Xiong, 2013;Li et al, 2011), but their evolution is not driven by the internal states of the social agents.…”

“…The previous social interaction models (Deffuant et al, 2000;Javarone & Squartini, 2014;Li et al, 2012;Chau et al, 2014;Das, Gollapudi & Munagala, 2014;Fang, Zhang & Thalmann, 2013;Li et al, 2011) do not assign nodes (i.e., individuals or social agents) the basic properties of humans, i.e., humans evolve, learn, react, and adapt in time. The reason for the simplicity behind the existing models is twofold: first, the state-of-the-art Opinion representation types where the larger nodes (labeled with S) represent stubborn agents (or opinion sources) which can also have any value for opinion, with the property that their opinion value never changes.…”

Modeling and Analyzing Engagement in Social Network Challenges

“…Several works focused on analyzing and mining users behavior on social networks [21,4,8]. We are not aware of work specifically dedicated to the engagement is social network challenges; we therefore describe social engagement in other contexts.…”

Sentiment Analysis in Spanish for Improvement of Products and Services: A Deep Learning Approach

“…Twitter is a social network, where users share information on almost everything in real time. Therefore, companies consider this social network as a rich source of information that allows knowing the general opinion about their products and services, among others [1]. However, analyzing and processing all these opinions require much time and effort for the humans.…”