超文本全文检索技术的研究与实现

摘要

在分析超文本全文检索功能需求的基础上，提出了一种将非结构化的超文本转换成为结构化数据库，以支持全文检索的实现思想和模型，形式化地描述了超文本检索引擎的原理，并分析了模型设计与实现的关键技术问题，如超文本关键词提取、查询条件构造、全文检索算法及查询结果处理等．该模型具有实现简单、通用性强、支持大容量超文本信息检索等特点，并已在鲁迅图书馆数字化网络化信息系统和钱学森特色数据库系统中得到了验证和应用．%Base on the analysis of the requirements and functions of the full-text search engine for hypertext, a new total solution model which can support the hypertext full-text searching is presented. The kernel of this model can convert the unstructured document (hypertext) to the well structural hypertext database. Moreover, the principle of the full-text search engine is described formally, and some key issues in the implementation of the search engine such as key word pick-up, condition express construction, full-text search algorithm etc. are also analyzed. The model presented is characterized by its simplicity, generality and the ability of search for large hypertext repository, and it is also verified and applied in the Lu Xun Digital Library System and Qian Xuesen Digital Library System.