Linking Data Across Universities: an integrated video lectures dataset
Nov 25, 2011 2687 views 
This paper presents our work and experience interlinking educational information across universities through the use of Linked Data principles and technologies. More specifically this paper is focused on selecting, extracting, structuring and interlinking information of video lectures produced by 27 different educational institutions. For this purpose, selected information from several websites and YouTube channels have been scraped and structured according to well-known vocabularies, like FOAF, or the W3C Ontology for Media Resources. To integrate this information, the extracted videos have been categorized under a common classification space, the taxonomy defined by the Open Directory Project. An evaluation of this categorization process has been conducted obtaining a 98% degree of coverage and 89% degree of correctness. As a result of this process a new Linked Data dataset has been released containing more than 14,000 video lectures from 27 different institutions and categorized under a common classification scheme.