Analysis of Academic Success Using Machine Learning: Addiction and ChatGPT

• Autores: Torres-Diaz, Juan Carlos, Reátegui Rojas, Ruth María
• Formato: artículo
• Fecha de Publicación: 2024
• Institución: Universidad de Lima
• Repositorio: Revistas - Universidad de Lima
• Lenguaje: español
• OAI Identifier: oai:revistas.ulima.edu.pe:article/7390
• Enlace del recurso: https://revistas.ulima.edu.pe/index.php/Interfases/article/view/7390
• Nivel de acceso: acceso abierto
• Materia: ChatGPT; addiction; machine learning; adicción; aprendizaje automático

This paper analyzes the impact of the variables phone addiction, pornography addiction, number of times the phone is unlocked per hour, and level of confidence in ChatGPT on the academic success of a group of 4278 students from eight universities in Ecuador. The decision trees (DT), random forest (RF), and support vector machine (SVM) methods are used. The results obtained indicate similar levels of precision achieved in the three algorithms; in terms of accuracy, in the case of SMOTE, DT is the algorithm that presents the highest accuracy (accuracy = 0,64); and, in the case of RandomOverSampler, the SVM algorithm had the highest accuracy (accuracy = 0,59).