Harvesting Metadata from Open Educational Resources for Semantic Annotation of Online Educational Content

Abstract

The massive and ever increasing amount of information on the web has made it difficult to perform a faster and more relevant "search and discovery" for educational content. The current keyword-based search model fails to take into consideration the relevance of an educational resource from various perspectives. Tagging the online educational resources with metadata allows much faster and more accurate searching. The Semantic Web community and the Learning Resource Metadata Initiative (LRMI) have come up with a list of properties suitable for tagging educational resources. However, the challenge is the absence of a vocabulary for possible values of LRMI-recommended properties. In this work, we propose a method for building an educational vocabulary for online learning resources such as OERs, by harvesting and collating metadata from multiple open educational contents. The novelty of this method, over other works, lies in using metadata from educational resources already tagged by education community, which is indicative of the usefulness of the metadata. We further propose a semi-automatic framework for tagging online educational resources with values for the LRMI-recommended properties. This provides an innovative tool that can be used by educators and students alike, for creating and consuming learning resources enhanced with metadata, for mass adoption of OER // Paper ID 140