-
Notifications
You must be signed in to change notification settings - Fork 19
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge pull request #20 from sgonzalezsilot/main
Participación charla santiago_gonzalez.md
- Loading branch information
Showing
1 changed file
with
17 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,17 @@ | ||
# Using NLP, Emotions and Explainability for Detecting Depression | ||
|
||
## Authors | ||
Santiago González Silot, Universidad de Jaén; Eugenio Martínez-Cámara, Universidad de Jaén; Luis Alfonso Ureña-López, Universidad de Jaén. | ||
|
||
## Presenter | ||
Santiago González Silot, PhD Student at University of Jaen | ||
|
||
## Abstract | ||
Mental health is becoming increasingly important in society as mental disorders have devastating consequences for patients, their families and their social circle. | ||
Preventive detection of mental disorders, and in particular depression, can help in their medical treatment. Therefore, this presentation explains the participation of the winning team (SINAI, UJA) in the early detection of mental disorders risk in Spanish competition [MentalRisk](https://sites.google.com/view/mentalriskes). | ||
We will explore the different techniques used to try to mitigate some of the problems of LLMs and their use in such complex tasks, we will detail the 2 systems developed, which are based on the use of a detection model of the 6 basic emotions of Ekman together with advanced Fine-Tuning techniques to adapt the model to the context and finally we will analyze these systems using SHAP to know what the model pays attention to in order to make a decision and detect possible errors and biases. | ||
|
||
## Resumen | ||
La salud mental cada vez es más importante en la sociedad ya que los trastornos mentales provocan consecuencias devastadores en los pacientes, sus familias y su circulo social. | ||
La detección preventiva de los trastornos mentales, y en particular de la depresión, puede ayudar en su tratamiento médico. Por ello, en esta presentación se explica la participación del equipo ganador (SINAI, UJA) en la competición de detección temprana de la depresión en español [MentalRisk](https://sites.google.com/view/mentalriskes). | ||
Se explorarán las distintas técnicas usadas para intentar paliar algunos de los problemas de los LLMs y su uso en temáticas tareas tan complejas, se detallarán los 2 sistemas desarrollados, los cuales están basados en el uso de un modelo de detección de las 6 emociones básicas de Ekman junto a técnicas avanzadas de Fine-Tuning para adaptar el modelo al contexto y finalmente se analizarán estos sistemas usando SHAP para así poder conocer en que presta atención el modelo para tomar una decisión y detectar posibles errores y sesgos. |