林业企业黄页Deep Web数据集成研究

发布时间：2018-05-15 08:01

本文选题：林业企业黄页 + Deep　；参考：《北京林业大学》2012年硕士论文

【摘要】：由于互联网技术的飞速发展,Web信息急剧膨胀,传统搜索引擎注重查全率而查准率不足,导致Web信息搜索越来越困难。特定领域的Deep Web网站上蕴藏着丰富的主题信息,但是这些Deep Web网站通常是分散的、不易检索的,因此,Deep Web数据集成技术逐渐成为研究热点。它不但可以解决传统搜索引擎不能索引Deep Web数据的局限,还能把各个孤立的Deep Web数据集成到一起,提供面向主题的一站式搜索服务。林业Deep Web情况亦然,用户难以利用传统的技术全面快速地收集Deep Web中的企业信息,使用Deep Web数据集成技术对林业企业黄页进行集成,提供高效专业化的Web信息服务,能够让用户更快的找到相关信息,不仅促进林业信息化发展,同时,也有利于推动林业经济的发展。本文通过对林业企业黄页Deep Web特征的分析,提出根据层次结构制定导航规则的Deep Web爬虫模型,设计出数据抽取的模板。同时,本文还对用户身份验证、导航URL提取、翻页URL提取、页面属性字段的提取、数据去重等Deep Web数据集成关键技术做了探索性的研究,通过实验对各种方案进行了尝试和比较,然后给出了最优的解决方案,并最终开发出高可用性的林业企业黄页Deep Web数据集成系统。最后,本文对系统的改进做了一些展望,为进一步的研究提供了基础。
[Abstract]:Due to the rapid development of Internet technology and the rapid expansion of web information, the traditional search engines pay more attention to recall and insufficient precision, which makes it more and more difficult to search for Web information. There are abundant subject information on Deep Web websites in specific fields, but these Deep Web websites are usually scattered and difficult to retrieve. Therefore, deep Web data integration technology is becoming a hot research topic. It can not only solve the limitation that traditional search engine can not index Deep Web data, but also integrate all isolated Deep Web data together to provide a one-stop search service for topic oriented. The same is true of forestry Deep Web. It is difficult for users to collect enterprise information in Deep Web by traditional technology, and to integrate yellow pages of forestry enterprises with Deep Web data integration technology to provide efficient and specialized Web information service. It not only promotes the development of forestry information, but also promotes the development of forestry economy. Based on the analysis of Deep Web features of yellow pages in forestry enterprises, this paper proposes a Deep Web crawler model for navigation rules based on hierarchical structure, and designs a template for data extraction. At the same time, the key technologies of Deep Web data integration, such as user authentication, navigation URL extraction, page turning URL extraction, page attribute field extraction, data de-reprocessing and so on, are also studied in this paper. This paper attempts and compares various schemes through experiments, then gives the optimal solution, and finally develops a high-availability yellow page Deep Web data integration system for forestry enterprises. Finally, this paper makes some prospects for the improvement of the system, which provides the basis for further research.
【学位授予单位】：北京林业大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP393.09;F326.2

【参考文献】