【作者】 张立；

【导师】 齐德昱； 罗健生；

【作者基本信息】 华南理工大学 ， 软件工程， 2011， 硕士

【摘要】 信息技术发展了近40年,其代表产物——信息管理系统的发展也经历了从对输入进行简单的运算、到可以积累数据从而形成规律和知识、再到可以支持人们决策的过程。然而,在信息管理系统的功能越来越强大的同时,系统中存放数据的形式也越来越复杂和多样化,系统的响应时间也越来越长。对于用户而言,他们越来越难在系统查询界面里得到自己想要的信息。这个时候,具备快速和全面的信息搜索工具——搜索引擎走上了历史的舞台。本文详细地介绍了具有全文索引功能的搜索引擎的特点、相关理论和关键技术,并对构建一个搜索引擎提出了一个合理而有效的设计方案。它讨论了具有全文索引功能的搜索引擎开发设计的全过程,包括:网站的选定,网页的抓取,对抓取结果的后期处理、数据存储,以及信息搜索的技术等。本文的突出之处在于在利用原有开源框架的基础上,对技术进行了集成和改进。其一、改进了爬虫系统的网页抓取算法,让抓取效率大大提高;其二、在数据存储和数据搜索方面,均采用了数据库和索引文件混合的方式,既提高了搜索的效率,又改善了数据存放的结构。本文的意旨除了开发一个文本搜索引擎外,还提出了一个能用于以往信息管理系统内提供数据搜索的思路和基本模型。本文所描述的系统具有以下特点:1、在JAVA跨平台技术和面向对象技术下实现了搜索引擎,具有很好的扩展性和代码重用性。2、基于索引文件为主体的存储方式,大大提高了用户使用搜索引擎查询的效率。3、通过对开源软件的研究,善用其作为本系统开发的辅助工具,从而提高了技术的先进性、结构的可靠性和开发的便利性。更多还原

【Abstract】 Development of information technology for nearly 40 years, which represents the product: MIS. The development of MIS has gone from a simple operation on the input to the formation of knowledge on the data accumulated, then people s decision can be support. However, the more powerful of MIS, the more complex and diversity for data storing, the more responsing time on system. For users, they are more and more difficult to get what they want from the system. At this time, search engine which is faster and more comprehensive for information searching is onto the historical stage. This paper introduces features, theories and key technologies of a full-text indexing search engine. And it present a rational and effective design for building a search engine.This article focused on the process of development and design of full-text indexing search engine, including: site selection, web crawling, crawling the results of post-processing, data storage, and information search technology, etc.This article highlights is that integration and improvement of technology based on the existing open-source framework. First, it improved algorithm for crawling the page of the crawler, so that the efficiency of capture was greatly improved. Second, in terms of data storage and data search, both the database and index files are used mixed together, so that it not only improves the efficiency of search, but also improves data storage structure.In addition to the intention of developing a text search engine, this paper also presents an idea and the basic model of providing data search in MIS which is in the past. This system has the following characteristics:1、Good scalability and code reusability benefit from the cross-platform technologies in JAVA and object-oriented technology.2、Greatly improving efficiency of quering from users benefit from the index files as mainly storage method.3、Researching on and good used of the open source software, impoved the advancement of technology, reliability of structure and convenience of development.更多还原