Semantic-Enhanced Recommendation of Video Lectures

Abstract

Learning support systems explore several audio-visual resources to consider individual needs and learning styles aiming to stimulate learning experiences. However, the large amount of educational content in different formats and the possibility of making them available in a fragmented way turns difficult the tasks of accessing these resources and understanding the concepts under study. Although literature has proposed approaches to explore explicit semantic representation through artifacts such as ontologies in learning support systems, this research line still requires further investigation efforts. In this paper, we propose a method for recommending educational content by exploring the use of semantic annotations over textual transcriptions from video lessons. Our investigation addresses the difficulties in extracting entities from natural language texts as subtitles of videos. We report on major challenges to achieve the representation of video transcriptions as semantic annotations for automatic recommendation of educational content