26th September 2023

MentalRiskES

Early detection of mental disorders risk in Spanish

Registration closed

 Description

According to a recent report by the World Health Organization, there is 1 in every 8 people in the world suffering from a mental disorder. The COVID-19 pandemic has raised the prevalence of anxiety and depression to more than 26% in just one year. Suicide is the fourth leading cause of death among 15-29 year-olds. The organisation considers that early identification is a key effective intervention to prevent these problems.

Consequently, there is a growing interest in detecting and identifying mental disorders in social media streams. This answers a demand from society due to the high increase in these problems among the population, in several kinds of mental risks: eating disorders, dysthymia, anxiety, depression, suicidal ideation, and others. Actually, relevant evaluation campaigns like the Cross-Lingual Evaluation Forum (CLEF)  have hosted during the last years the Early-Risk Identification task (eRisk). Unfortunately, these campaigns have focused mainly on English, leaving aside other languages, like Spanish. 

MentalRiskEs is a novel task on early risk identification of mental disorders in Spanish comments from Telegram users. The task must be resolved as an online problem, that is, the participants must be able to detect a potential risk as early as possible in a continuous stream of data. Therefore, the performance not only depends on the accuracy of the systems but also on how fast the problem is detected. These dynamics are reflected in the design of the tasks and the metrics used to evaluate participants. For this first edition, the disorders considered are eating disorders, depression, and an unknown one which is intended to evaluate the robustness of approaches for new disorders not known a priori.

If you want to participate in the MentalRiskEs@IberLEF2023 shared task, please fill this form. Once you are registered, you can ask any questions through the Google Group of the shared task MentalRiskEs@IberLEF2023.

Participants will be required to submit their runs and their submissions will be evaluated on the test partitions for the corresponding corpora. The task is designed as an online problem, so participants will be instructed on the usage of a submission API to get the next message in the history of the user and to submit a prediction for that user according to the timeline of messages retrieved so far. Again, a base notebook showing how to perform retrieval and submissions will be delivered to participants. A phase with trial data is foreseen to check systems before the evaluation period.

Participants will be required to submit their runs and are asked to describe their systems in paper submissions. We encourage participating teams to highlight the real contribution of their systems in identifying successful approaches along with failed attempts and findings on how to advance in more performant solutions. This description must contain the following details:

 Contact

If you have any specific question about the MentalRiskEs 2023 task, we may ask you to let us know through the Google Group MentalRiskEs@IberLEF2023.

For any other questions that do not directly concern the shared task, please contact with Arturo Montejo Ráez or Flor Miriam Plaza-del-Arco.

Arturo 

Montejo-Ráez

  University of Jaén 

amontejo@ujaen.es

Email

Flor Miriam 

Plaza-del-Arco 

 Bocconi University 

flor.plaza@unibocconi.it

Email

MentalRiskES at IberLEF2023

SINAI Research Group

Twitter: @NLP_SINAI

This work has been partially supported by WeLee project (1380939, FEDER Andalucía 2014-2020) funded by the Andalusian Regional Government, and projects CONSENSO (PID2021-122263OB-C21), MODERATES (TED2021-130145B-I00), SocialTOX (PDC2022-133146-C21) funded by Plan Nacional I+D+i from the Spanish Government, and project PRECOM (SUBV-00016) funded by the Ministry of Consumer Affairs of the Spanish Government.