Twitter Dataset for Mental Disorders Detection

 We provide two datasets extracted from Twitter, in Spanish and English, and annotate each one with approximately 1,500 users who have been diagnosed with one of nine different mental disorders (ADHD, Autism, Anxiety, Bipolar, Depression, Eating disoders, OCD, PTSD and Schizophrenia) along with 1,700 matched-control users. For both datasets, the outcome is a total of just over 3,000 Twitter users with their corresponding timelines (the texts retrieved from each user cover at least 3 months of activity on the social media), which support two user-level classification tasks, binary and multiclass.

 — Dataset usage terms : By using this dataset, you agree to (i) use the content of this dataset and the data generated from the content of this dataset for non-commercial research only, and (ii) remain in compliance with Twitter's Developer Policy.