Computational analysis and automated classification of eating disorder discourse on Twitter’s EDTWT community
Jace Li, Chuduo Zhang, Anam Nawaz Khan, Hollie Raynor, Xueping Li · 2026
Read Computational analysis and automated classification of eating disorder discourse on Twitter’s EDTWT community by Jace Li, Chuduo Zhang, Anam Nawaz Khan, Hollie Raynor, Xueping Li (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13505132.txt.
Description
Jace Li and co-authors analysed eating-disorder discussion on Twitter, focusing on the community known as EDTWT. They gathered 48,341 tweets from 18,587 users posted between January 2022 and February 2025. Their aims were to map posting habits, feelings, and themes, and to build tools that sort such posts automatically.
A small share of users produced nearly half the posts, and activity peaked late on Friday nights. About a quarter of tweets used clinical keywords, covering body image, restricted eating, and recovery. Topic modelling found ten themes, including calorie tracking and Spanish- and Polish-language subgroups. An ensemble classifier sorted posts with strong accuracy, beating a biomedical language model on rare categories. The authors say such tools could support platform moderation and earlier help for users at risk.
Catalog record
| CREATOR | Jace Li, Chuduo Zhang, Anam Nawaz Khan, Hollie Raynor, Xueping Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56093 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |