Loading...
Please wait, while we are loading the content...
Similar Documents
Graduate School Query Reformulation in Trec a Thesis Submitted to the Faculty of the Graduate School of the University of Minnesota by Yunjiang Luo in Partial Fulfillment of the Requirements for the Degree of Master of Science
| Content Provider | Semantic Scholar |
|---|---|
| Author | Crouch, Carolyn J. |
| Copyright Year | 2006 |
| Abstract | This is to certify that I have examined this copy of a master's thesis by YUNJIANG LUO and have found that it is completed and satisfactory in all respects, and that any and all revisions required by the final examining committee have been made. ABSTRACT An information retrieval system based on the vector space model selects documents from a database in response to a user's query and ranks these documents according to their correlation with the query. Query formulation is a significant factor in the successful search of a document collection. Relevance feedback, which is based on the automatic reformulation of the query after the user has evaluated a small subset of the returned documents for relevance, is a method known to provide significant improvement in terms of retrieval effectiveness. The purpose of this research is to study the effectiveness of query reformulation using different feedback methods on the TREC (Text Retrieval Conference) collection of some 1,000,000+ documents. All of this work is an attempt to verify the results produced by Singhal [9]. Singhal used the small, standard Smart collection for his experiments. This work uses the TREC collection and attempts to verify that the conclusions reached by Singhal are valid in this very large database environment. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.d.umn.edu/cs/thesis/yunjiang_luo_ms.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
