Detección y análisis de patrones en datos académicos para la predicción de deserción estudiantil

Abstract

Abstract: This Curricular Integration Project (CIP) aims to predict the dropout rate of university students based on the analysis of various personal, academic, and interaction patterns. The CRISP-DM methodology is used to develop this CIP, as it comprises a series of phases that allow for a concrete approach to addressing the problem of student dropout. In addition, several data preprocessing techniques are used, as well as the implementation of machine learning algorithms, which allow for the identification of various relevant patterns that influence student behavior. The main contribution of CIP is to optimize data analysis processes, demonstrating the potential of various technologies to address these problems in a scalable, ethical, and sustainable manner, resulting in better-informed decision-making and the effective design of intervention strategies that promote student retention by higher education institutions.