Graph-based semantic annotation for enriching educational content with linked data

摘要

In this paper, a new approach to semantic annotation with linked data in the field of document enrichment is presented. This application has been developed in the domain of Education and contrary to traditional semantic annotation, which relates each relevant term of the document with an instance of the ontology, in our approach relevant terms are connected to a (sub)graph of the ontology. Specifically, each relevant term is related to an instance which is expanded to a predefined depth limit, so the term is finally annotated with a (sub)graph. During the expansion process, instances unrelated with the document topics are ruled out, so only relevant and contextualized information is finally included. As result of this process, the document is annotated with a set of interconnected (sub)graphs, and students may access and navigate through these contents to deepen the topics described in the document. This approach has several benefits. From the document enrichment perspective, a set of (sub)graphs, provides a better description, moreover considering the semantic nature of linked data. From the computational perspective, this approach is also more suitable, particularly in the domain of Education. Filtering linked data is computationally expensive, and thus this process cannot be performed in real time. Our approach has been validated in the e-Leaming domain and compared with similar approaches that also annotate with linked data.