Abstract
This paper presents an approach for an automated discovery, categorization and retrieval of metadata enriched E-learning resources. This research has further been implemented on a real platform “HyperManyMedia” at Western Kentucky University. It also presents the contribution that this work has made to the Adaptive Hypermedia systems. The main research question guiding this work is whether it is feasible and benecial to add the metadata to learning resources, while still being able to retrieve learning resources that are satisfactory and effective for the learner. The main goals of our HyperManyMedia system as implementation based on the suggested architecture, are to: (1) provide the learner with metadata search engine to overcome the limitations of searching for learning objects, and (2) be an open source repository for online learning.