Journals Proceedings

Abstract

Wikiversity is an online open-source public repository for University-level teaching and learning materials, based on the Wikipedia architecture for “crowd-sourcing”: it relies on volunteers to collaborate by actively contributing their knowledge for the common good. Undoubtedly a huge volume of learning resources exists on the WWW, but scattered on individual websites, in a wide range of formats and structures. Individual lecturers prepare online teaching materials to support their own teaching, but few know of Wikiversity, and few have time or inclination to take on the extra step of formally registering and uploading their materials to Wikiversity. In this paper we will present an approach to organise and semi-automate the harvesting of these scattered resources, by using BootCat to crawl the Web and adapting Web-as- Corpus techniques. Results show that scouting the Web returns better resources than BootCat toolkit which returns a lot of irrelevant links such as forums, and personal repositories.