Aprendizaje automático en el diagnóstico de la autoestima y depresión en el ámbito de la salud mental
Fecha
2025
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Nacional San Luis Gonzaga.
Resumen
Objetivo: Evaluar la eficiencia del aprendizaje automático en el diagnóstico de la autoestima y
depresión en el ámbito de la salud mental. Método: el método empleado fue aplicado del nivel
aplicativo; se aplicó los cuestionarios de Rosemberg para la autoestima y el cuestionario de Beck para
la depresión a una muestra de 167 jóvenes para recoger los datos, se realizó un análisis descriptivo
de los datos para conocer el estado de la depresión y autoestima, al mismo tiempo los datos fueron
sometidos al modelo de inteligencia artificial con el modelo de aprendizaje supervisado con el
algoritmo SVM, el modelo de inteligencia artificial fue desarrollado con la herramienta Orange
Datamining. Se desarrolló un asistente virtual con inteligencia artificial de apoyo para jóvenes con
estado depresivo o de autoestima baja Resultado: los resultados del algoritmo para la depresión
arrojaron una precisión del 79,3%, mientras que para los datos de autoestima mostraron una precisión
del 95.9%. La estadística descriptiva de los datos los jóvenes arrojaron más de 20% con estado
depresivo entre extremo, grave e intermitente; para el caso de la autoestima arrojó datos muy críticos
con más del 50% con autoestima baja y media. Conclusión: se concluye la importancia del modelo
de predicción de salud mental, la estadística que refleja el estado de salud de los jóvenes relacionados
con la depresión y la autoestima, se concluye y el desarrollo de un asistente virtual de apoyo.
Objective: To evaluate the efficiency of machine learning in the diagnosis of self-esteem and depression in the field of mental health. Method: the method used was applied at the applicative level; the Rosemberg questionnaires for self-esteem and the Beck questionnaire for depression were applied to a sample of 167 young people to collect data, a descriptive analysis of the data was carried out to know the state of depression and self-esteem, at the same time the data were submitted to the artificial intelligence model with the supervised learning model with the SVM algorithm, the artificial intelligence model was developed with the Orange Datamining tool. A virtual assistant with artificial intelligence was developed to support young people with depressive state or low self-esteem. Result: the results of the algorithm for depression showed an accuracy of 79.3%, while for self-esteem data showed an accuracy of 95.9%. The descriptive statistics of the data showed more than 20% of young people with extreme, severe and intermittent depression; in the case of self-esteem, the data was very critical, with more than 50% with low and medium self-esteem. Conclusion: we conclude the importance of the mental health prediction model, the statistics reflecting the health status of young people related to depression and self-esteem, and the development of a virtual support assistant.
Objective: To evaluate the efficiency of machine learning in the diagnosis of self-esteem and depression in the field of mental health. Method: the method used was applied at the applicative level; the Rosemberg questionnaires for self-esteem and the Beck questionnaire for depression were applied to a sample of 167 young people to collect data, a descriptive analysis of the data was carried out to know the state of depression and self-esteem, at the same time the data were submitted to the artificial intelligence model with the supervised learning model with the SVM algorithm, the artificial intelligence model was developed with the Orange Datamining tool. A virtual assistant with artificial intelligence was developed to support young people with depressive state or low self-esteem. Result: the results of the algorithm for depression showed an accuracy of 79.3%, while for self-esteem data showed an accuracy of 95.9%. The descriptive statistics of the data showed more than 20% of young people with extreme, severe and intermittent depression; in the case of self-esteem, the data was very critical, with more than 50% with low and medium self-esteem. Conclusion: we conclude the importance of the mental health prediction model, the statistics reflecting the health status of young people related to depression and self-esteem, and the development of a virtual support assistant.
Descripción
Palabras clave
Salud mental, Depresión, Autoestima, Inteligencia artificial, Asistente virtual, Mental health
