_____________________________________________________
Dataset for depression detection in Spanish Language
_____________________________________________________
by LTL-INAOE 2019 for research/academic purposes


This is a collection of posts in Spanish Language from the Twitter platform for the depression detection task. The users  were  labeled  as  depressed  if  any  of  their  posts  matched with these expressions :"Me diagnosticaron/He sido diagnosticado con /Me han diagnosticado"... "depresión". On the other hand, users were labeled as non-depressive if any of their posts contained the string  “depresión”.

The folder contains the data collection organized in two files:

- tweets_Español_depresivos.json: contains a series of tweets of depressive users.
- tweets_Español_no-depresivos.json: stores a collection of tweets of non-depressive users.

Every line of the files corresponds to the information for each user: user name and a list of tweets. Each tweet includes: date, id, and the message.


References:

---"Crosslingual depression detection in Twitter using bilingual word alignments." Laritza Coello-Guilarte, Rosa María Ortega-Mendoza, Luis Villaseñor-Pineda, Manuel Montes-y-Gómez. CLEF 2019
