Show simple item record Ibrahim, Duraid M en_US 2007-06-01T19:18:55Z 2007-06-01T19:18:55Z 2000-01-01T00:00:00Z en_US
dc.description.abstract Query formulation and expansion have long been explored for enhancing query effectiveness and solving the word mismatch problem in information retrieval systems. Most of the approaches are statistical in nature. They are based on the occurrence frequency of words this thesis, we present a new approach based on natural language processing. Given a natural language query, our approach will translate a natural language query into a Boolean query that is better, in terms of retrieval effectiveness, than the original query. The terms in the Boolean query are assigned weights based on their contribution to the semantic of the query, which is determined by its occurrence frequency and its syntactic dependency within the query. Furthermore, the resulting weighted Boolean query can be further improved by expanding the query terms with synonyms in a very restrictive fashion. This process is fully automated and does not require human intervention. Experiments run for TREC-4 queries showed consistent improvement over standard information retrieval ranking methods. en_US
dc.format.extent 4594991 bytes
dc.format.extent 184 bytes
dc.format.mimetype application/pdf
dc.format.mimetype text/plain
dc.language en en_US
dc.language.iso en_US
dc.rights info:eu-repo/semantics/openAccess
dc.title Natural language query translation and expansion in information retrieval en_US
dc.type info:eu-repo/semantics/masterThesis Computer Science en_US Master of Science (M.Sc.) en_US

Files in this item

This item appears in the following Collection(s)

Show simple item record

View Statistics