Classifier Model For Personalizing Exercises Given To Students Using Artificial Neural Networks; [使用人工神经网络向学生提供的个性化练习的分类器模型]; [Классификационная модель для индивидуализации упражнений, предлагаемых студентам, с применением искусственных нейронных сетей]

Peralta E.A.S.; Allazo E.A.V. · Publicaciones de la Facultad de Educacion y Humanidades del Campus de Melilla (2023) · DOI: 10.30827/publicaciones.v53i2.26818

Resumen

The objective of the article is to develop a model that allows to predict in a personalized way the exercises that a student can solve, and on the other hand the exercises that a student cannot solve. The model is based on a series of factors that influence students’ learning rates. The course that was used as an experiment in the project is the handling of functions in spreadsheets. For the development of the process, the KDD (Knowledge Discovery in Databases) data mining methodology has been used and artificial neural networks with backward learning (Backpropagation), which is a supervised learning algorithm, have been used for the model. The model is fed with data such as gender, age, academic grade, parents’ level of education, type of school, previous grades of the subjects that the student obtains while advancing in the course. The research approach is quantitative, experimental, applied and the population was represented by eighty five students. The result shows that the model achieves a 72% probability of accuracy in predicting the assignment of exercises to students according to their characteristics. Exercises that cannot be solved are given a help for their better understanding and resolution.

Instituciones: Universidad Nacional de San Agustín de Arequipa; Universidad Tecnológica del Perú