Although arguable success of today’s keyword based search engines in certain information retrieval tasks, ranking search results in a meaningful way remains an open problem. In this work, the goal is to use of semantic relationships for ranking documents without relying on the existence of any specific structure in a document or links between documents. Instead, real-world entities are identified and the relevance of documents is determined using relationships that are known to exist between the entities in a populated ontology. We introduce a measure of relevance that is based on traversal and the semantics of relationships that link entities in an ontology. We expect that the semantic relationship-based ranking approach will be either an alternative or a complement to widely deployed document search for finding highly relevant documents that traditional syntactic and statistical techniques cannot find.
Available at: http://works.bepress.com/amit_sheth/173/
Paper presented at the IEEE Fourth International Conference on Semantic Computing, Pittsburgh, PA, September 22-24, 2010.