Desarrollo de un framework que identifica, describe y organiza recursos educativos disponibles en plataformas web de universidades mediante minería de datos.

Abstract

This Master's thesis intends to analyze the degree of adoption of the semantic web with the Schema.org vocabulary and the JSON-LD format by universities to publish and describe educational resources on their websites. It is based on the fusion of the methodologies: Design Science Research (DSR) to guide the creation of an artifact and CRISP-DM to address the data mining process. Since these methodologies are not exclusive, it was possible to relate the various phases that compose them [1]. The result of this work is a framework that consists of eleven components distributed in three well-defined layers: data access layer, service layer and application layer. The components are developed throughout the combined phases of the two methodologies and address the data mining process from download to final analysis of the data to obtain information. The results show that universities do use the Schema vocabulary and the JSON-LD format to publish resources, although not to the extent that would be expected. In addition, the absence of properties of the Schema vocabulary that allow a specific definition of a resource in an educational context was noted. The final part of this work proposes some actions to extend the functionality of the framework to other contexts different from the educational one, as well as other vocabularies or formats.