Local variation of hashtag spike trains and popularity in Twitter

TL;DR

This paper introduces a method to analyze hashtag activity on Twitter by comparing it to neuron spike trains, revealing that popular hashtags tend to have more regular activity patterns, which could help predict online popularity.

Contribution

It applies local variation analysis to hashtag time series, demonstrating its effectiveness in distinguishing between popular and less popular hashtags based on their dynamic patterns.

Findings

Popular hashtags are less bursty and more regular.

The methodology can differentiate between different levels of hashtag popularity.

Potential for predicting online popularity based on activity patterns.

Abstract

We draw a parallel between hashtag time series and neuron spike trains. In each case, the process presents complex dynamic patterns including temporal correlations, burstiness, and all other types of nonstationarity. We propose the adoption of the so-called local variation in order to uncover salient dynamics, while properly detrending for the time-dependent features of a signal. The methodology is tested on both real and randomized hashtag spike trains, and identifies that popular hashtags present regular and so less bursty behavior, suggesting its potential use for predicting online popularity in social media.