The corpus contains Instagram messages from users suffering from mental disorders. There is a sample of messages per user. The annotations are made at message level and indicate whether the user is suffering from depression or not.
Language(s)
Spanish
Year
2023
Domain
Health
Text types
Telegram messages
Format
json
Data access
Registration
Publication
Alba María Mármol-Romero, Adrián Moreno-Muñoz, Flor Miriam Plaza-del-Arco, María Dolores Molina-González, Maria Teresa Martín-Valdivia, Luis Alfonso Ureña-López, Arturo Montejo-Raéz (2023) Overview of MentalRiskES at IberLEF 2023: Early Detection of Mental Disorders Risk in Spanish. Procesamiento del Lenguaje Natural, Revista nº 71, septiembre de 2023, pp. 329-35
NLP Topic
Number of units
12036
Training set size
6248
Test set size
5164
Development set size
624
Size - additional information
binary label indicating whether the author of the message suffers from depression or not and four multiclass labels indicating whether the author suffers from depression, is against or in favour the disorder or is a control subject.
