Towards data-driven educational management: a platform for analyzing student dropout rates at CUJAE

Abstract

This work addresses the transition to data-driven educational management through the development and implementation of a digital platform designed to optimize the analysis and monitoring of student dropouts at CUJAE. The solution replaces the traditional system—based on physical forms and spreadsheets, prone to errors and inefficient—with a modern application that operates offline and synchronizes later. The platform integrates artificial intelligence techniques and non-parametric models to analyze the causes and patterns of dropout, enabling the generation of real-time statistics. The methodology followed comprised: (1) diagnosis of the manual process, (2) selection of an open technology stack (Vue.js, NestJS, SQLite, Prisma, Capacitor), (3) analysis and design of the solution, (4) implementation of the solution, and (5) validation with real data. The results validated a significant reduction in errors, an improvement in operational efficiency, and an increase in institutional analytical capacity. The solution not only modernizes an administrative process, but also constitutes a strategic advance by providing timely and reliable information for evidence-based decision-making and the development of more effective educational policies.