高级检索|图表检索

图书情报工作 ›› 2013, Vol. 57 ›› Issue (20): 123-126.DOI: 10.7536/j.issn.0252-3116.2013.20.021

• 知识组织 • 上一篇    下一篇

一种实用型跨语言检索查询翻译接口的设计与实现

高影繁, 徐红姣   

  1. 中国科学技术信息研究所
  • 收稿日期:2013-08-18 修回日期:2013-09-13 出版日期:2013-10-20 发布日期:2013-10-20
  • 作者简介:高影繁,中国科学技术信息研究所助理研究员,博士,E-mail:gaoyingf@istic.ac.cn;徐红姣,中国科学技术信息研究所实习研究员,硕士.
  • 基金资助:

    本文系中国科学技术信息研究所重点工作课题"多语言科技信息语义关联网络构建及其应用"(项目编号:ZD2012-3-3)研究成果之一。

Design and Implementation of a Practical Query Translation Interface for Cross-language Information Retrieval

Gao Yingfan, Xu Hongjiao   

  1. Institute of Scientific and Technical Information of China, Beijing 100038
  • Received:2013-08-18 Revised:2013-09-13 Online:2013-10-20 Published:2013-10-20

PDF

摘要：

面对日益膨胀的多语种信息资源,跨语言信息检索已成为实现全球知识存取和共享的关键技术手段。构建一个实用型的跨语言检索查询翻译接口,可方便地嵌入任意的信息检索平台,扩展现有信息检索平台的多语言信息处理能力。该查询翻译接口采用基于最长短语、查询分类和概率词典等多种翻译消歧策略,并从查询翻译的准确性和接口的运行效率两个角度对构建的查询翻译接口进行评测,实验结果验证所采用方法具有可行性。

关键词: 查询翻译消歧, 查询翻译接口效率, 评测

Abstract:

With the huge multilingual information resources, there's an obvious language barrier for the existing information retrieval platform to provide documents organization and management services. A practical query translation interface for Cross-Language Information Retrieval (CLIR) has been established in the paper. It could be embedded expediently in any information retrieval platform, expanding the multilingual information processing ability for the existing system. A hybrid corpora-based method for translation disambiguation is proposed in the paper. Provided an English-Chinese bilingual dictionary, parallel corpus have been employed to append translation probability value to bilingual dictionary. Moreover, maximum forward matching of phrases and query classification method were also used. The test collection has been used in testing CLIR systems successfully, and the experimental results demonstrate the methods of query translation and translation disambiguation for CLIR in the paper is feasible.

Key words: query translation disambiguation, efficiency of query translation interface, evaluation

中图分类号: