Eating disorders (EDs) are serious mental illnesses associated with physical and psychiatric problems, and premature death. Examining social media communication about ED symptoms may provide insight into how to prevent and treat these disorders. This study is to explore topics on Twitter related to EDs. We applied the Correlation Explanation (CorEx) topic model on 18,288 ED-related tweets and identified 20 topics, which were further grouped into 8 categories. The top two topic categories are body image and ED consequences. We manually evaluated the relevance of tweets to their assigned topics and average accuracy is 77.86%. Our findings are consistent with another study using content analysis, and we identified additional topics, such as ED consequences, pornography, and treatment and education from these tweets.
|Original language||English (US)|
|Title of host publication||2019 IEEE International Conference on Healthcare Informatics, ICHI 2019|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|State||Published - Jun 2019|
|Event||7th IEEE International Conference on Healthcare Informatics, ICHI 2019 - Xi'an, China|
Duration: Jun 10 2019 → Jun 13 2019
|Name||2019 IEEE International Conference on Healthcare Informatics, ICHI 2019|
|Conference||7th IEEE International Conference on Healthcare Informatics, ICHI 2019|
|Period||6/10/19 → 6/13/19|
Bibliographical notePublisher Copyright:
© 2019 IEEE.
- Body image
- Eating disorder
- Social media
- Topic model