Information retrieval on the Semantic Web: Integrating inference and retrieval

vision of the Semantic Web is that it will be much like the Web we know today, except that documents will be enriched by annotations in machine understandable markup. These annotations will provide metadata about the documents as well as machine interpretable statements capturing some of the meaning of document content. We discuss how the informa- tion retrieval paradigm might be recast in such an environ- ment. We suggest that retrieval can be tightly bound to infer- ence. Doing so makes today's Web search engines useful to Semantic Web inference engines, and causes improvements in either retrieval or inference to lead directly to improvements in the other.