Análisis exploratorio y visualización de patrones en datos académicos usando técnicas de minería de datos

Abstract

Abstract: This study focuses on the application of data mining techniques to analyze academic data related to Information Technology students. Through meticulous data cleaning, normalization, and imputation processes, data quality and consistency were enhanced. The K-means algorithm was used in the CRISP-DM methodology to identify patterns, revealing significant insights into student academic performance, age-related grade distribution, the impact of scholarships, and variations across cities and academic semesters. Emphasizing the importance of code refactoring and diverse data visualization techniques, this research underscores the value of a comprehensive approach. The findings affirm K-means' effectiveness in identifying coherent clusters, and the combined data cleansing and data mining approach offers a holistic and valuable perspective on academic data.