自然语言检索的实现及其关键问题

The Pivotal Issues and Implementation of Natural Language Information Retrieval

在网络环境下,非专业人员的网络信息检索活动的比重大幅度提高,而这些用户大多没有经过检索培训,没有使用检索词构造提问式、制定检索策略的经验,这样,自然语言检索技术的研究与应用日益受到信息检索界的高度重视。自然语言检索包括自然语言标引和自然语言提问处理两方面的内容,自然语言检索的关键问题基本是围绕这两个方面的,这些问题包括面向自然语言检索的标引技术、自然语言提问分析与处理、自然语言检索的匹配过程及概念控制等问题。本文介绍了自然语言检索的研究情况,讨论了自然语言检索的相关问题。

Under web environment, more and more non- professionals have the chance to engage themselves in various activities over the Interact, while most of them are poor of searching knowledge. It is not an easy thing to set a proper seeking strategy for them, such as to choose the correct keywords to build a correct query. So the study on natural language information retrieval （NLR） and its application have attracted high attentions of reseachers in the field of information retrieval increasingly. NLR consists of two parts at least, which are both associated and relativly independent with each other： indexing of natural language documents and processing of natural language queries. Almost all pivotal issues in NLR around those two parts, which includes documents indexing, natural language queries analying and processing, matching process and cncept controling. This paper discusses the progress of the NLR and some related questions.