首页> 外国专利> QUERY TRANSLATOR AND METHOD FOR CROSS-LANGUAGE INFORMATION RETRIEVAL USING LIGUISTIC RESOURCES FROM WIKIPEDIA AND PARALLEL CORPUS

QUERY TRANSLATOR AND METHOD FOR CROSS-LANGUAGE INFORMATION RETRIEVAL USING LIGUISTIC RESOURCES FROM WIKIPEDIA AND PARALLEL CORPUS

机译:使用维基百科和并行语料库中的合法资源进行跨语言信息检索的查询翻译器和方法

摘要

The present invention relates to a query translation system and method for a cross-language information search engine using linguistic resources from Wikipedia and a parallel corpus, which can easily establish a multilingual dictionary from Wikipedia and increase the performance of translation by using a small parallel corpus. The query translation system comprises: a query input module in which a query expressed in a source language is inputted; a query division module for dividing the inputted query expressed in the source language into translatable units; a query translation module for translating the query or extracting a translation candidate by using a set of the established linguistic resources from Wikipedia, with the divided query; a query selection module for eliminating translation ambiguity through the operation of calculating a meaning relationship between queries which are immediately translated or have a translation candidate and then selecting the best translated word; and a query output module for outputting the query to perform search with the query expressed in a final target language.
机译:本发明涉及一种使用来自维基百科和并行语料库的语言资源的跨语言信息搜索引擎的查询翻译系统和方法,其可以容易地从维基百科建立多语言词典并通过使用小的并行语料库来提高翻译性能。 。查询翻译系统包括:查询输入模块,其中输入以源语言表达的查询;以及查询划分模块,用于将输入的源语言表达的查询划分为可翻译单元;查询翻译模块,用于通过使用来自维基百科的一组已建立的语言资源来翻译查询或提取翻译候选,并进行划分查询;查询选择模块,用于通过计算立即翻译或具有翻译候选的查询之间的含义关系,然后选择最佳翻译词来消除翻译歧义;查询输出模块,用于输出查询以执行以最终目标语言表达的查询的查询。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号