Analytical study of quality-biased competition dynamics for memes in social media

TL;DR

This paper presents an analytical model for understanding how memes of varying quality compete and spread in social media, providing insights into their lifetime and popularity distributions.

Contribution

It introduces an analytical framework for meme competition dynamics that accounts for quality bias, enabling better understanding and control of information spread.

Findings

Derived lifetime and popularity distributions of memes.

Mapped meme dynamics to a solvable diffusion process.

Reproduced stationary properties with a mean-field approach.

Abstract

The spreading of news, memes and other pieces of information occurring via online social platforms has a strong and growing impact on our modern societies, with enormous consequences, that may be beneficial but also catastrophic. In this work we consider a recently introduced model for information diffusion in social media taking explicitly into account the competition of a large number of items of diverse quality. We map the meme dynamics onto a one-dimensional diffusion process that we solve analytically, deriving the lifetime and popularity distributions of individual memes. We also present a mean-field type of approach that reproduces the average stationary properties of the dynamics. In this way we understand and control the role of the different ingredients of the model, opening the path for the inclusion of additional, more realistic, features.