#greysanatomy vs. #yankees: Demographics and Hashtag Use on Twitter

TL;DR

This study infers demographic information from Twitter profile images to analyze hashtag usage patterns among different groups, revealing both shared and group-specific hashtags over a one-year period.

Contribution

It introduces a method to infer demographics from profile images and applies it to analyze hashtag use across demographic groups on Twitter.

Findings

Overlap in popular hashtags across groups

Presence of group-specific hashtags

Demographics influence hashtag usage patterns

Abstract

Demographics, in particular, gender, age, and race, are a key predictor of human behavior. Despite the significant effect that demographics plays, most scientific studies using online social media do not consider this factor, mainly due to the lack of such information. In this work, we use state-of-the-art face analysis software to infer gender, age, and race from profile images of 350K Twitter users from New York. For the period from November 1, 2014 to October 31, 2015, we study which hashtags are used by different demographic groups. Though we find considerable overlap for the most popular hashtags, there are also many group-specific hashtags.