The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
A ambiguidade semântica é um problema sério na recuperação de informação. A expansão da consulta foi proposta como um método para resolver este problema. Entretanto, as consultas tendem a não ter muitas informações para ajustar os vetores de consulta à semântica latente, que são difíceis de expressar em poucos termos de consulta fornecidos pelos usuários. Propomos um método de modificação de vetores de documentos que modifica vetores de documentos com base na relevância dos documentos. Espera-se que este método mostre melhor eficácia de recuperação do que os métodos convencionais. Neste artigo, avaliamos nosso método por meio de experimentos de recuperação nos quais é avaliada a relevância de documentos extraídos de artigos científicos e uma comparação com tf
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Teruhito KANAZAWA, Atsuhiro TAKASU, Jun ADACHI, "A Relevance-Based Superimposition Model for Effective Information Retrieval" in IEICE TRANSACTIONS on Information,
vol. E83-D, no. 12, pp. 2152-2160, December 2000, doi: .
Abstract: Semantic ambiguity is a serious problem in information retrieval. Query expansion has been proposed as one method of solving this problem. However, queries tend not to have much information for fitting query vectors to the latent semantics, which are difficult to express in a few query terms given by users. We propose a document vector modification method that modifies document vectors based on the relevance of documents. This method is expected to show better retrieval effectiveness than conventional methods. In this paper, we evaluate our method through retrieval experiments in which the relevance of documents extracted from scientific papers is assessed, and a comparison with tf
URL: https://global.ieice.org/en_transactions/information/10.1587/e83-d_12_2152/_p
Copiar
@ARTICLE{e83-d_12_2152,
author={Teruhito KANAZAWA, Atsuhiro TAKASU, Jun ADACHI, },
journal={IEICE TRANSACTIONS on Information},
title={A Relevance-Based Superimposition Model for Effective Information Retrieval},
year={2000},
volume={E83-D},
number={12},
pages={2152-2160},
abstract={Semantic ambiguity is a serious problem in information retrieval. Query expansion has been proposed as one method of solving this problem. However, queries tend not to have much information for fitting query vectors to the latent semantics, which are difficult to express in a few query terms given by users. We propose a document vector modification method that modifies document vectors based on the relevance of documents. This method is expected to show better retrieval effectiveness than conventional methods. In this paper, we evaluate our method through retrieval experiments in which the relevance of documents extracted from scientific papers is assessed, and a comparison with tf
keywords={},
doi={},
ISSN={},
month={December},}
Copiar
TY - JOUR
TI - A Relevance-Based Superimposition Model for Effective Information Retrieval
T2 - IEICE TRANSACTIONS on Information
SP - 2152
EP - 2160
AU - Teruhito KANAZAWA
AU - Atsuhiro TAKASU
AU - Jun ADACHI
PY - 2000
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E83-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2000
AB - Semantic ambiguity is a serious problem in information retrieval. Query expansion has been proposed as one method of solving this problem. However, queries tend not to have much information for fitting query vectors to the latent semantics, which are difficult to express in a few query terms given by users. We propose a document vector modification method that modifies document vectors based on the relevance of documents. This method is expected to show better retrieval effectiveness than conventional methods. In this paper, we evaluate our method through retrieval experiments in which the relevance of documents extracted from scientific papers is assessed, and a comparison with tf
ER -