面向房产领域的垂直搜索引擎研究与实现

发布时间：2018-11-06 08:00

【摘要】：伴随着互联网的高速发展,网络信息呈指数级增长。在如此海量的信息当中需要搜索引擎定位所需的信息。虽然通用搜索引擎在一定程度上能解决资源定位问题,但是其效果并不理想,对于专业领域的信息检索,很难达到用户的搜索需求。垂直搜索引擎的出现就是为了解决通用搜索引擎在专业领域的不足,其对特定领域信息的深度挖掘弥补了通用搜索引擎信息泛而不精的缺点。论文从理论和实践两个方面对垂直搜索引擎中关键技术做了深入研究。本文首先介绍了研究背景及意义、搜索引擎的分类和垂直搜索引擎在国内外的发展现状。其次介绍了垂直搜索引擎的基本工作原理、系统构成和关键技术。接下来,详细介绍了网页的主题表示,构造了本文的主题特征向量,分析了主题网页的分布特征。深入研究了基于内容的主题相关度判定和基于链接结构的主题相关度判定,分析了各自的缺点和不足。在基于内容的主题相关度判定的基础上引入网页重要度,设计了一个基于网页内容和基于网页链接结构的主题爬虫算法。对于主题爬虫中的主题孤岛问题,设计了一个基于动态调整最大深度的隧道穿越算法,该算法一定程度上缓解了网络孤岛问题。随后,设计了一个基于房产领域的垂直搜索引擎,对本系统进行了系统分析,设计了本系统的总体框架,详细介绍了各个子功能模块的设计与实现,并对本系统做了性能分析和功能测试。最后,对论文所做的工作进行相应的总结,并且提出需要进一步研究的工作。
[Abstract]:With the rapid development of the Internet, network information increases exponentially. In such a large amount of information needs search engine location needs information. Although the general search engine can solve the problem of resource location to a certain extent, its effect is not ideal, and it is difficult to reach the user's search demand for the information retrieval in the specialized field. The emergence of vertical search engine is to solve the shortcomings of general search engine in the professional field, and its deep mining of information in a specific field makes up for the shortcomings of general search engine information. In this paper, the key technologies of vertical search engine are studied in theory and practice. This paper first introduces the research background and significance, the classification of search engines and the development of vertical search engines at home and abroad. Secondly, the basic working principle, system structure and key technology of vertical search engine are introduced. Then, the theme representation of the web page is introduced in detail, the theme feature vector is constructed, and the distribution feature of the theme page is analyzed. In this paper, the content based topic correlation decision and the link structure based topic relevance judgment are studied in depth, and their shortcomings and shortcomings are analyzed. A topic crawler algorithm based on web content and web link structure is designed by introducing the importance of web pages on the basis of content-based topic correlation judgment. For the topic isolated island problem in the topic crawler, a tunnel crossing algorithm based on dynamic adjustment of maximum depth is designed, which to some extent alleviates the problem of network islanding. Then, a vertical search engine based on the real estate field is designed, the system is systematically analyzed, the overall framework of the system is designed, and the design and implementation of each sub-function module are introduced in detail. The performance analysis and function test of the system are also done. Finally, the work of the paper is summarized, and further research work is proposed.
【学位授予单位】：南昌大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP391.3

【参考文献】