Towards a statistical physics of dating apps

TL;DR

This paper models the distribution of likes in dating apps using statistical physics, revealing how user popularity evolves and identifying a condensate phenomenon where few users dominate in likes.

Contribution

It introduces a novel application of stochastic and coagulation process methods to analyze dating app dynamics and demonstrates the emergence of a condensate in user popularity.

Findings

Likes distribute unevenly among users.

A condensate forms with a small fraction of users receiving most likes.

Elo-based rating models show gelation-like behavior.

Abstract

Over the last ten years, a sharp rise in the number of dating apps has broadened the spectrum of how one can get in contact with new acquaintances. A common feature of such apps is a swipe enabling a user to decide whether to like or dislike another user. As is the case in real life, a user may be more or less popular, which implies the distribution of likes among different users is not trivial. In this paper we show how likes are distributed across users, based on different decision-making strategies and on different app settings. We apply theoretical methods originally developed in stochastic and coagulation processes to the investigation into the dynamics of dating app networks. More specifically, we show that whenever a dating app differentially displays different users with respect to their popularity in different models, users are split into two categories: a first category including users who have received most likes and a second category, referred to as a condensate, which in the long-term will be reduced to a small fraction of likes or to no likes at all. Finally, we explore different models based on a rating system of the users, known as Elo. These models will turn out to exhibit behaviour typical of gelating systems and non-trivial distribution of Elo rating.