Randy Goebel, Sandra Zilles, Christoph Ringlstetter, Andreas Dengel, Gunnar Grimnes
The World Wide Web Consortium (W3C) has been the consolidator for many ideas regarding the evolution of the World Wide Web (WWW), including the promotion of the so-called Semantic Layer Cake model for the development of the semantic web. The semantic layer cake provides a framework to discuss a variety of approaches to an integrated view of the meta-data that will support a broad range of machine and human manipulation of digital information. Our goal is to develop a deeper understanding of the potential role of the semantic layer cake, by investigating some of the detailed relationships between the components. We also attempt to articulate some of the issues regarding the development of the semantic layer cake for real application. The path to this goal covers many fundamental issues underlying all of knowledge representation research. Paramount in this discussion is the goal of identifying the potential role of machine learning within the semantic layer cake, and finding the tradeoffs between the extremes of a completely automated construction of the semantic web via machine learning, versus one that is completely hand-engineered using the tools emerging from each layer of the semantic layer cake.
Subjects: 1.10 Information Retrieval; 11. Knowledge Representation
Submitted: Jan 26, 2008