Análisis de rasgos lingüísticos con técnicas de procesamiento del lenguaje natural en la detección temprana de depresión
The development of computational methods using information from the Web for early detection of risks is a socially relevant, scientifically attractive and currently a growing area of research. Depression is one of the most frequent mental disorders in the world and with high incidence of suicide i...
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Online |
| Lenguaje: | spa |
| Publicado: |
Instituto de Lingüística, Facultad de Filosofía y Letras, Universidad Nacional de Cuyo
2021
|
| Materias: | |
| Acceso en línea: | https://revistas.uncu.edu.ar/ojs3/index.php/analeslinguistica/article/view/5522 |
| Sumario: | The development of computational methods using information from the Web for early detection of risks is a socially relevant, scientifically attractive and currently a growing area of research. Depression is one of the most frequent mental disorders in the world and with high incidence of suicide in the most severe cases. Therefore, early detection of this illness could lead to a timely treatment and to save lives. This paper analyzes the relationship between computational models that allow the automatic detection of depression and the linguistic properties of the text written by people who experience the disease. State-of-the-art text representations in document classification are used, covering linguistic, syntactic and semantic aspects. The results obtained with standard classifiers indicate that word embeddings capture precise information to detect quickly and safely signs of depression. |
|---|
