Passage-Level Document Retrieval Using Lexical Chains
-
- MOCHIZUKI Hajime
- School of Information Science, Japan Advanced Institute of Science and Technology
-
- IWAYAMA Makoto
- Advanced Research Lab., Hitachi Ltd.
-
- OKUMURA Manabu
- School of Information Science, Japan Advanced Institute of Science and Technology
Bibliographic Information
- Other Title
-
- 語彙的連鎖に基づくパッセージ検索
Search this article
Description
The importance of document retrieval systems which can retrieve relevant documents for user's needs is now increasing with the growing availability of full-text documents. The passage-level document retrieval has been received much attentions in the recent document retrieval task. The definition of a passage is considered as a sequent part of document which contain a relating content to a content of a query. In the passage retrieval, it is a problem how to decide the passages. It is considered that the passages which form meaning cohenrent units are effective in the improvement of the accuracy. Furthermore, it is also effective that the size and location of each passage is calculated flexibly with each query and document. In this paper we describe a definition of a passage calculation which can be able to calculate a similarity between each content of user's query and each part of document, using lexical chains which denote local document contexts. We also present a passage-level document retrieval method which improve the accuracy.
Journal
-
- IPSJ SIG Notes
-
IPSJ SIG Notes 127 39-46, 1998-09-17
Information Processing Society of Japan (IPSJ)
- Tweet
Details 詳細情報について
-
- CRID
- 1571417127177356416
-
- NII Article ID
- 110002934760
-
- NII Book ID
- AN10115061
-
- Text Lang
- ja
-
- Data Source
-
- CiNii Articles