An approach for semantic-based searching in learning resources

Abstract

Currently, online learning has been widely applied in education and training. Especially, when it is difficult for lecturers and learners to get close to each other in the context of Covid-19 epidemic period, online learning shows its availability and necessary. Learning materials provided in the educational institutions are diverse; almost lectures are stored as files but have not been totally arranged in a standard database system. Therefore, searching information about curriculum and lectures still face difficulties. This paper proposes a solution for semantic-based searching in learning resources. Firstly, ontologies are built to represent information of lectures. When users enter a query, the system pre-processes it (e.g., word segmentation, removing stop words), and then provides it to classifier (e.g., SVM) to identify the corresponding domain (or topic), aiming to narrow the search space in the ontology. After classifying, the key phrases will be queried in the appropriate ontology to result in related lectures. Experiments on lectures in the domains of information technology show that the proposed model is feasible.