IDENTIFYING INFLUENTIAL FACTORS IN STUDENT DROPOUT USING DECISION TREES

This document presents the development of a classification model to analyze the factors that influence a student at the Universidad Politécnica Salesiana to drop out of their degree program. This analysis is based on data provided by the university. The approach is based on classifications using decision trees. The methodology follows the Knowledge Discovery in Databases (KDD) process and consists of five steps: selection, processing, transformation, data mining, and evaluation. Using Python's Classification and Regression Tree (CART) algorithm, a tree with five levels and seventeen rules was created to identify potential dropouts. It concludes that factors such as the level of studies, academic performance, and the number of subjects taken by the student in a term are decisive in the decision to drop out.