| Presentation | 2010-06-28 Two step adjustment technique of term weight Hiroya YANO, Tai NAKAJIMA, Hayato YAMANA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | TF・IDF method is one of the methods to weight terms in the field of document retrieval. IDF value shows the degree of how a term is difficult to appear in the document set, and depends on the document set to be retrieved. Therefore, the problem is that, even if a term is difficult to appear in the same field of document set as query(which means the term is highly specific in the document), IDF value of term which appears easily in the document set to be retrieved is small. In this paper, we propose and study two step adjustment technique of term weight. In the first step, we get documents related to query using vector space model. In the next step, we retrieve relevant documents using IDF calculated from the document set acquired in the first step. Experiments using NTCIR-1 IR task collection indicate that, the precision of proposed method is improved about 7.1 percent comparing to that vector space model, and is almost the same value of the precision which get the highest in NTCIR-1. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | document retrieval / TF・IDF / pseudo-relevance feedback |
| Paper # | DE2010-9 |
| Date of Issue |
| Conference Information | |
| Committee | DE |
|---|---|
| Conference Date | 2010/6/21(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Data Engineering (DE) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Two step adjustment technique of term weight |
| Sub Title (in English) | |
| Keyword(1) | document retrieval |
| Keyword(2) | TF・IDF |
| Keyword(3) | pseudo-relevance feedback |
| 1st Author's Name | Hiroya YANO |
| 1st Author's Affiliation | Graduate School of Fundamental Science and Engineering, Waseda University() |
| 2nd Author's Name | Tai NAKAJIMA |
| 2nd Author's Affiliation | Graduate School of Fundamental Science and Engineering, Waseda University |
| 3rd Author's Name | Hayato YAMANA |
| 3rd Author's Affiliation | Faculty of Science and Engineering, Waseda University:National Institute of Informatics |
| Date | 2010-06-28 |
| Paper # | DE2010-9 |
| Volume (vol) | vol.110 |
| Number (no) | 107 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |