Internet comments as a barometer of public opinion

TL;DR

This paper investigates how Internet comments reflect public opinion and proposes a method to predict social responses based on comment voting patterns, extending existing opinion dynamics models with real-world case analyses.

Contribution

It introduces a new approach to estimate community responses to opinions using comment voting data, enhancing models of opinion dynamics with practical Internet data.

Findings

Internet comment votes correlate with social influence levels

Comment response patterns can predict public opinion shifts

Case studies demonstrate the method's predictive potential

Abstract

Social susceptibility is defined and analyzed using data from CNN news website. The current models of opinion dynamics, voting, and herding in closed communities are extended, and the community's response to the injection of a group with predetermined and permanent opinions is calculated. A method to estimate the values of possible response in Internet communities that follow a specific developing subject is developed. The level of social influence in a community follows from the statistics of responses ("like" and "dislike" votes) to the comments written by the members of the same community. Three real cases of developing news stories are analyzed. We suggest that Internet comments may predict the level of social response similar to a barometer that predicts the intensity of a coming storm in still calm environment.