Análisis de Causas de Abandono Estudiantil a través de Técnicas de Machine Learning en la UPS sede Guayaquil

Abstract

This study investigated the causes of student dropout at the Universidad Politécnica Salesiana in Guayaquil, using machine learning techniques to analyze survey data. Key factors such as gender, academic level, career perception, motivation, economic difficulties, and mental health were identified as significant in predicting student dropout. The Random Forest algorithm proved to be very effective in this analysis. Despite limitations related to the sample size, this study provides valuable information for developing intervention policies focused on improving student retention. The importance of addressing the economic and psychological well-being of students is emphasized, and future research is suggested to further explore the identified factors and develop effective strategies to prevent student dropout.