Based on current Web collaborative annotation systems, the paper first analyzes the common features of tags in documents as well as context-awareness to extend the concept of tags and make mapping between ontology, then uses ontology reasoning to enrich the semantics of tags, to capture implied semantic information, and to find out the relationships between documents, it finally detects pseudo-relevance among different documents in order to improve the accuracy of knowledge retrieval and knowledge recommendation as well as the flow of knowledge among subjects.
展开▼