Chatbot for the Detection of Depression in University Students in Peru using Sentiment Analysis

Saldana Condori W.A.; Mori Leon A.D.; Dextre Alarcon J.S. · Proceedings of the 2024 IEEE 4th International Conference on Advanced Learning Technologies for Education and Research, ICALTER 2024 (2024) · DOI: 10.1109/icalter65499.2024.10819245

Resumen

Mental disorders, such as depression, usually manifest themselves on average by the age of 24, when many young people are in college. During the COVID-19 pandemic, university students experienced a significant impact on their mental health due to confinement, which highlights the need for and importance of early detection of depression so that it can be treated effectively and in a timely manner. Because of this, a chatbot based on sentiment analysis is proposed for the early detection of depression in university students. The algorithm used in the chatbot makes use of responses to the PHQ-9 questionnaire modified to open-ended questions to predict the level of depression that the student may have. This proposal has 4 stages: Technology Analysis, Algorithm Development, Results and Chatbot Construction. The results obtained showed that the Gradient Boosting machine learning technique obtained an accuracy of 0.81. This result suggests that our chatbot can effectively and early detect depression in university students, providing a valuable contribution to the mental health care of students.

Instituciones: Universidad Peruana de Ciencias Aplicadas