Ontology-Based Knowledge Management

The Web's very popularity is making it more difficult to find, present, and maintain the data that users with a wide range of tasks and computer skills need. Existing document management systems use keyword matching as a search method, combined with information retrieval rather than query answering. In addition, these systems offer limited information-sharing facilities, and they do not support different views on documents or information maintenance. To address these weaknesses, a European consortium formed the On-to-Knowledge Project to build an ontology-based tool suite that efficiently processes the many heterogeneous, distributed, and semistructured documents typically found in intranets and on the Web. The consortium's approach integrates Semantic Web search technology, document exchange via transformation operators, automated information extraction, and systematic support for information maintenance and user-specific views. The paper considers how On-to-Knowledge's tools exploit the power of ontologies to provide automated support for acquiring, maintaining, and accessing weakly structured information sources.