Zhejiang University computer science professor De-Ren Chen and PhD student Wen-Ying Guo detail an e-learning implementation strategy using Semantic Web technology to imbue e-learning with more flexibility. The authors write that using ontologies to describe learning materials could allow the lack of a shared comprehension between terms in one vocabulary and between terms in various metadata vocabularies to be avoided. Their approach involves the provision of metadata for describing the content, context, and structure of learning materials, and Chen and Guo note that one of the most commonly used metadata schemes currently on the Web is the Dublin Core Metadata Initiative's Dublin Core Schema. Dublin Core is designed for metadata of all categories of digital resources, and therefore does not abide by the specific requirements the authors cite in describing learning resources. Thus, an extension of Dublin Core called the Learning Objects Metadata Standard (LOM) was created by the IEEE's Learning Technology Standards Committee, and LOM allows each learning object to be represented via a series of 70-plus attributes split up into nine categories. Chen and Guo define two ontology classes: An application ontology that describes the individual who wants to choose the course to study, and one that specifies training domain providers, including courses, location, and time. Two basic operations--semantic querying and semantic mapping--are employed to achieve semantic solution. The approach is designed to address several problems that can crop up in an e-learning environment: The expression of semantically identical concepts by dissimilar terms in the domain vocabulary, and the use by two applications of the same term with different definitions
