Multi-label Search Model for Open Educational Resources Based on Learning Purpose

Villalba-Condori K.; Tejada-Ortega L.; Vera-Sancho J.; Mamani-Calcina J.; Vera-Vasquez C.; Cardona-Reyes H. · Communications in Computer and Information Science (2022) · DOI: 10.1007/978-3-031-24709-5_10

Resumen

The Open Educational Resources (OER) repositories store a large quantity and variety of data from multiple sources, and that present relevant and useful information for the student training process, improving student performance thanks to more up-to-date and didactic educational content, this generates the need for a more specialized Learning Objects (LO) search system. In this research work, a multi-label search platform is proposed based on the learning purpose that unifies the information from different OER in a single platform, using an information extraction process based on Web Scraping techniques and generating a selection of one of them. Or multiple tags, based on the SOLO taxonomy, using an MLP-based multi-tag classifier. As a result, we obtained the improvement of the most specialized search and content according to what teachers really need, up to 85% accuracy of the resources they needed and approval of the use of the platform up to 78.57%.

Instituciones: Universidad Tecnológica del Perú; Universidad Continental; Universidad Católica de Santa María