SAmgI: Automatic Metadata Generation v2.0
Purchase or Subscription required for access
Purchase individual articles and papers
Subscribe for faster access!
Subscribe and receive access to 100,000+ documents, for only $19/month (or $150/year).
Already have access?
Institutional Subscription
You don't appear to be accessing the site through a subscribing institution (your IP address is 54.225.1.66).
If your university, college, or library subscribes to LearnTechLib, you may be able access full text articles through a login page.
You can search for your instition by name or by location.
Authors
EdMedia + Innovate Learning, Jun 25, 2007 in Vancouver, Canada ISBN 978-1-880094-62-4
Abstract
Manual creation of metadata is tedious, error-prone and doesn't scale. As the amount of learning content continues to grow, it becomes less and less feasible to describe all available content manually. Moreover, the metadata humans create are not perfect. Therefore, we need a change in approach, trying to automate this process as much as possible. In this paper, we provide an overview of our work from the last three years on how to achieve this automation. We also present a software design and implementation that has been evaluated in practice and that is being deployed at large scale. We detail the architectural design goals that made us evolve from a first approach to the current one.
Citation
Meire, M., Ochoa, X. & Duval, E. (2007). SAmgI: Automatic Metadata Generation v2.0. In C. Montgomerie & J. Seale (Eds.), Proceedings of ED-MEDIA 2007--World Conference on Educational Multimedia, Hypermedia & Telecommunications (pp. 1195-1204). Vancouver, Canada: Association for the Advancement of Computing in Education (AACE). Retrieved March 19, 2024 from https://www.learntechlib.org/p/25528.
© 2007 AACE