Thesis Creación, análisis y optimización de un cuestionario de salud mental mediante Inteligencia Artificial
No Thumbnail Available
Date
2025-12
Authors
Pérez Cerda, Yasmine Victoria
Journal Title
Journal ISSN
Volume Title
Program
Ingeniería Civil Informática
Departament
Campus
Campus Santiago San Joaquín
Abstract
La salud mental en estudiantes universitarios es una problemática creciente, especialmente en relación con trastornos como el burnout, la ansiedad y la depresión, que afectan tanto el bienestar emocional como el rendimiento académico. Actualmente, la evaluación de la salud mental en estudiantes se realiza mediante cuestionarios, entendidos como formularios que integran uno o más instrumentos psicométricos validados. Entre los más utilizados se encuentran el PHQ-9, GAD-7, WHOQOL-BREF y EUBE, los cuales permiten diagnosticar síntomas de depresión, ansiedad y burnout. No obstante, la aplicación conjunta de estos instrumentos genera un cuestionario extenso y complejo, lo que disminuye la cantidad de los datos y la participación estudiantil. Es por ello, que este trabajo tiene como objetivo analizar dichos instrumentos y las respuestas recopiladas de un estudio aplicado en la UTFSM, con el fin de explorar la creación de un cuestionario unificado y optimizado mediante técnicas de inteligencia artificial. A través del análisis de los ítems de los instrumentos, la identificación, de redundancias y correlaciones, y el uso de algoritmos de IA, se busca diseñar una versión más breve y eficiente, capaz de apoyar la detección temprana y facilitar la intervención en estudiantes de riesgo de desarrollar trastornos emocionales.
Mental health among university students is a growing concern, particularly regarding disorders such as burnout, anxiety, and depression, which negatively affect both emotional well-being and academic performance. Currently, the assessment of student mental health is carried out through questionnaires, understood as forms that integrate one or more validated psychometric instruments. Among the most widely used are PHQ-9, GAD-7, WHOQOL-BREF, and EUBE, which enable the diagnosis of symptoms of depression, anxiety, and burnout. However, the joint application of these instruments results in a lengthy and complex questionnaire, which decreases data quantity and student participation. Therefore, this work aims to analyze these instruments and the responses collected from a study conducted at Universidad Técnica Federico Santa María, with the purpose of exploring the development of a unified and optimized questionnaire through artificial intelligence techniques. By analyzing instrument items, identifying redundancies and correlations, and applying AI algorithms, the goal is to design a shorter and more efficient version, capable of supporting early detection and facilitating intervention in students at risk of developing emotional disorders.
Mental health among university students is a growing concern, particularly regarding disorders such as burnout, anxiety, and depression, which negatively affect both emotional well-being and academic performance. Currently, the assessment of student mental health is carried out through questionnaires, understood as forms that integrate one or more validated psychometric instruments. Among the most widely used are PHQ-9, GAD-7, WHOQOL-BREF, and EUBE, which enable the diagnosis of symptoms of depression, anxiety, and burnout. However, the joint application of these instruments results in a lengthy and complex questionnaire, which decreases data quantity and student participation. Therefore, this work aims to analyze these instruments and the responses collected from a study conducted at Universidad Técnica Federico Santa María, with the purpose of exploring the development of a unified and optimized questionnaire through artificial intelligence techniques. By analyzing instrument items, identifying redundancies and correlations, and applying AI algorithms, the goal is to design a shorter and more efficient version, capable of supporting early detection and facilitating intervention in students at risk of developing emotional disorders.
Description
Keywords
Salud mental, Estudiantes universitarios, Inteligencia artificial, Cuestionarios psicométricos, Detección temprana, Reducción de dimensionalidad, Análisis semántico
