Bladimir Cedeño Salazar, Efraín Díaz-Macías, Emma Yolanda Mendoza Vargas, Harold Elbert Escobar Terán
Abstract: Student dropout is a key challenge for higher education due to its academic, social, and institutional impacts. This study analyzes this issue at the Quevedo State Technical University (UTEQ) through a systematic literature review aimed at identifying the main associated factors and examining the potential of artificial intelligence to strengthen academic monitoring. The methodology was developed in accordance with PRISMA guidelines, ensuring transparency in the search, selection, and evaluation of studies. Of the more than 500 publications initially identified, 20 were included based on their thematic and methodological relevance. The results showed that socioeconomic and academic factors were the most influential, each accounting for 30% of the total. Furthermore, Colombia accounted for the largest volume of scientific output, followed by Peru, Mexico, and Chile. The study proposes incorporating machine learning techniques to detect at-risk students early, personalize interventions, and improve retention, performance, and educational continuity at UTEQ.