Metodología de detección temprana de ansiedad y depresión en jóvenes de Barranquilla basado en Machine Learning (ML) y datos socioculturales
No hay miniatura disponible
Fecha
2025
Autores
Peralta Fontalvo, Yisel Paola
Gnecco Ramos, Miguel Mauricio
Giraldo Rojas, Maria Fernanda
Arenas Barrera, Oscar Giovanny
Torres Gazabon, Julio Alberto
Título de la revista
ISSN de la revista
Título del volumen
Editor
Ediciones Universidad Simón Bolívar
Facultad de Ingenierías
Facultad de Ingenierías
Resumen
La salud mental de los jóvenes es un tema prioritario en la agenda educativa y social
de Barranquilla, dada la creciente prevalencia de síntomas de ansiedad y depresión
en contextos escolares. Este trabajo de grado propone una metodología de
detección temprana de ansiedad y depresión en adolescentes, combinando
técnicas de Machine Learning (ML) con el análisis de datos socioculturales, como
una herramienta preventiva adaptada al contexto local.
El estudio se enmarca en un enfoque mixto, que integra una fase cuantitativa
basada en la aplicación de modelos predictivos como Random Forest sobre datos
obtenidos de encuestas aplicadas a estudiantes y personal institucional, y una fase
cualitativa para validar e interpretar los hallazgos desde una perspectiva contextual
y profesional.
Se utilizó muestreo intencional, incluyendo a 35 estudiantes del Distrito de
Barranquilla, 3 psicólogas, 1 Directivo Docente. Las variables analizadas incluyen
condiciones familiares, académicas y emocionales, cruzadas con factores sociales
como redes de apoyo, acceso a recursos y nivel socioeconómico.
Los resultados permiten identificar patrones asociados al riesgo de ansiedad y
depresión, reforzados por el juicio experto de profesionales. Se proponen
lineamientos para implementar esta metodología como parte de una política
educativa preventiva y contextualizada, sin necesidad de desarrollar una aplicación
informática.
Mental health among young people is a priority issue on Barranquilla’s educational and social agenda, given the rising prevalence of anxiety and depression symptoms in school contexts. This thesis proposes a methodology for the early detection of anxiety and depression in adolescents, combining Machine Learning (ML) techniques with the analysis of sociocultural data as a preventive tool adapted to the local context. The study adopts a mixed-methods approach, integrating a quantitative phase based on the application of predictive models such as Random Forest to data obtained from surveys conducted with students and institutional personnel, and a qualitative phase aimed at validating and interpreting the findings from a contextual and professional perspective. Intentional sampling was used, involving 35 students from Barranquilla’s public education system, 3 psychologists, 1 educational coordinator. The variables analyzed include family, academic, and emotional conditions, along with social factors such as support networks, resource access, and socioeconomic level. The results reveal patterns associated with the risk of anxiety and depression, supported by expert professional judgment. Guidelines are proposed for implementing this methodology as part of a preventive and contextualized educational policy, without the need to develop a software application.
Mental health among young people is a priority issue on Barranquilla’s educational and social agenda, given the rising prevalence of anxiety and depression symptoms in school contexts. This thesis proposes a methodology for the early detection of anxiety and depression in adolescents, combining Machine Learning (ML) techniques with the analysis of sociocultural data as a preventive tool adapted to the local context. The study adopts a mixed-methods approach, integrating a quantitative phase based on the application of predictive models such as Random Forest to data obtained from surveys conducted with students and institutional personnel, and a qualitative phase aimed at validating and interpreting the findings from a contextual and professional perspective. Intentional sampling was used, involving 35 students from Barranquilla’s public education system, 3 psychologists, 1 educational coordinator. The variables analyzed include family, academic, and emotional conditions, along with social factors such as support networks, resource access, and socioeconomic level. The results reveal patterns associated with the risk of anxiety and depression, supported by expert professional judgment. Guidelines are proposed for implementing this methodology as part of a preventive and contextualized educational policy, without the need to develop a software application.
Descripción
Palabras clave
Inteligencia artificial, Salud mental, Adolescencia, Machine learning, Factores Socioculturales