Data Mining for Curriculum Optimization in Smart Education Systems

Alanya-Beltran J.; Silva-Cueva J.; Niño-Cueva D.; Niño-Cueva M.; Mancco-Rivas F.; Martinez-Rojas F. · 2025 5th International Conference on Advancement in Electronics and Communication Engineering, AECE 2025 (2025) · DOI: 10.1109/aece67531.2025.11386711

Resumen

The development of digital technologies allowed collecting enormous amounts of educational data, which opens new opportunities in the optimization of curriculum data mining. The research paper examines how machine learning algorithms, Random Forest, Decision Trees, and Neural Networks can be applied on Kaggle based dataset to evaluate the effectiveness of curriculum in smart education systems. The findings indicate significant improvements in the performance, with Random Forest having the best prediction accuracy of 85.7%. There was a substantial academic improvement with a 9.41% and 9.89% increase in the scores of Mathematics and Computer Science respectively. The study highlights the potential of data mining in the area of curriculum design and offers a framework of intelligent curriculum development in the educational field.

Instituciones: Universidad Nacional Mayor de San Marcos; Universidad San Ignacio de Loyola; Universidad Nacional de Educación Enrique Guzmán y Valle