中文檢索評估系統可行性研究 | Proposing a Prototype for Chinese Corpus Test Collection

<p>頁次:68-87</p><p class="MsoNormal" style="margin: 0cm 0cm 0pt; mso-layout-grid-align: none;"><span style="font-family: "新細明體","serif"; mso-font-kerning: 0pt; mso-ascii-theme...

Full description

Bibliographic Details
Main Author: 吳美美 Mei-Mei Wu
Format: Article
Language:English
Published: National Taiwan Normal University 1999-04-01
Series:Tushuguanxue yu Zixun Kexue
Online Access:http://jlis.glis.ntnu.edu.tw/ojs/index.php/jlis/article/view/349
Description
Summary:<p>頁次:68-87</p><p class="MsoNormal" style="margin: 0cm 0cm 0pt; mso-layout-grid-align: none;"><span style="font-family: "新細明體","serif"; mso-font-kerning: 0pt; mso-ascii-theme-font: major-fareast; mso-fareast-theme-font: major-fareast; mso-hansi-theme-font: major-fareast;"><span style="font-size: small;">本研究主要在探討建置一個中文檢索評估系統的可行性。數位化成為必要的趨勢之後,世界上各重要語言都受到其使用國家的重視與贊助而開始進行研究。<span lang="EN-US">1992</span>年起美國國家標準暨技術局<span lang="EN-US">(NIST)</span>和美國國防部尖端研究企畫署的資訊科技室<span lang="EN-US">(DARPA)</span>閒始組織一系列的文獻檢索會議<span lang="EN-US">(Text Retrieval Conference, TREC)</span>,每年舉辦一次評比研討會議,至今已有<span lang="EN-US">TREC 9</span>的召開。日本在籌畫五年之後,亦在今年<span lang="EN-US">(1999)</span>開始推展日文的檢索評比研討活動。本文提出建置一個中文檢索評估系統可行性研究,並期望政府重視支持而能建設一個良好的中文檢索研究環境,以便鼓勵更多的研究者參與中文資訊檢索的研究。<span lang="EN-US"></span></span></span></p><p class="MsoNormal" style="margin: 0cm 0cm 0pt; mso-layout-grid-align: none;"><span style="mso-fareast-font-family: 新細明體; mso-fareast-theme-font: minor-fareast;" lang="EN-US"><span style="font-size: small;"><span style="font-family: Times New Roman;">Chinese language retrieval evaluation has caught the attention of researchers studying cross language retrieval techniques internationally. The importance of Chinese language in information retrieval ( IR) has been recognized, however, the evaluation tool and research environment is limited. Given the importance of Networked Information Retrieval (NIR) and the accelerating development of Digital Library (DL), a Chinese corpus test collection and a set of evaluation criteria for IR evaluation is desperately needed for the IR research and development community.</span></span></span></p><p class="MsoNormal" style="margin: 0cm 0cm 0pt; mso-layout-grid-align: none;"><span style="mso-fareast-font-family: 新細明體; mso-fareast-theme-font: minor-fareast;" lang="EN-US"><span style="font-size: small;"><span style="font-family: Times New Roman;">The purpose of this study is to propose a prototype of CTREC. Main tasks and tracks, test collections, topics, measurement procedures and evaluation criteria are proposed. Some solutions for implementing CTREC are also suggested.</span></span></span></p>
ISSN:0363-3640