面向分布式数据源的语义垂直搜索系统研究与实现

发布时间：2018-06-27 22:08

本文选题：垂直搜索 + 领域本体　；参考：《哈尔滨工业大学》2013年硕士论文

【摘要】：近年来，随着Internet技术的高速发展，Web信息爆炸性增长，互联网已成为海量数据空间。人们从这些海量数据中得到自己需要的信息变的越来越困难，此时搜索技术变的尤为重要。随着搜索引擎技术的高速发展，通用搜索引擎的功能变的日益强大，，这使得通用搜索引擎的结构也日益复杂，拥有庞大的索引数据和十分宽泛的主题，并且主题与主题之间无分类。而且通用搜索的搜索深度不够，查准率较低，时效性较差等这些局限性决定了通用搜索并不能满足特定用户的需求。由于人们对特定领域、特殊行业的个性化需求,人们迫切需要更有针对性、更具搜索深度的搜索引擎，在这种情况下，面向特定领域的垂直搜索技术应运而生。目前已有很多垂直搜索引擎，如生活搜索酷讯、旅游搜索去哪儿网等，但这些垂直搜索引擎仍采用基于关键词的搜索方式，无法满足用户语义级别的检索要求。为了研究和实现语义层面的垂直搜索，提高检索质量，依据课题背景和课题来源，本文主要研究分布式环境下语义垂直搜索技术和系统实现。主要包括研究领域本体的构建，基于领域本体的语义查询扩展方法，基于DOM树的信息抽取，采用最新索引Lucene4.x等技术实现语义级别的垂直搜索，并做出实验数据分析。本文研究和完成的主要工作有：（1）通过计算机安全领域信息的采集、过滤、归类等处理构建面向计算机安全领域的领域本体，由于计算机安全领域涉及广泛，出于时间上的考虑，本文针对计算机安全领域狭窄范围进行本体的构建。并对各类的属性和类之间的关系进行定义描述，实现计算机安全领域知识的表示和信息组织。（2）根据计算机安全领域独有的特点，使用基于规则的DOM树的信息抽取技术初步实现对计算机安全领域的相关信息的自动抽取，为将来本体的编辑和维护打下基础。（3）对本体中的实例、概念、对象属性、数据属性以及类与类之间的关系构建索引，利用索引完成对用户的查询请求的语义分析和语义扩展，实现语义级别的信息检索。（4）根据某所提供的语义词典和关系范畴构建概念词语层次体系，为语义查询扩展打下基础。本文设计并实现了基于计算机安全领域本体的语义垂直搜索服务。实验表明，本文设计语义垂直搜索系统能较好的解决目前基于关键字检索中存在的机械式搜索缺陷以及查询精度不够等的缺陷，说明基于领域本体的语义垂直搜索具有现实意义。
[Abstract]:In recent years, with the rapid development of Internet technology and the explosive growth of Web information, the Internet has become a massive data space. It is becoming more and more difficult for people to get the information they need from these massive data, and the search technology is becoming more and more important. With the rapid development of search engine technology, the function of general search engine becomes more and more powerful, which makes the structure of general search engine more and more complex, with huge index data and very broad theme. And there is no classification between themes and themes. And the limitations of general search, such as insufficient search depth, low precision rate and poor timeliness, determine that general search can not meet the needs of specific users. Because of the individualized demand of specific fields and industries, people urgently need search engines with more pertinence and search depth. In this case, the vertical search technology for specific fields emerges as the times require. At present, there are many vertical search engines, such as life search cool news, travel search where to go to the net, but these vertical search engines still use the search method based on keywords, which can not meet the users' semantic level of retrieval requirements. In order to study and realize the vertical search in semantic level and improve the retrieval quality, according to the background and source of the subject, this paper mainly studies the technology and system implementation of semantic vertical search in distributed environment. It mainly includes the construction of domain ontology, the extension method of semantic query based on domain ontology, information extraction based on Dom tree, vertical search of semantic level using the latest index Lucene 4.x, and the analysis of experimental data. The main works of this paper are as follows: (1) the domain ontology of computer security domain is constructed by collecting, filtering and classifying the information in computer security domain. For the sake of time, this paper constructs ontology for the narrow scope of computer security field. The definition and description of each kind of attribute and the relationship between classes are carried out to realize the representation and information organization of computer security domain knowledge. (2) according to the unique characteristics of computer security field, The information extraction technology based on rule based Dom tree is used to realize the automatic extraction of relevant information in computer security domain, which lays the foundation for the editing and maintenance of ontology in the future. (3) for the instance, concept, object attribute of ontology. The data attribute and the relation between class and class are used to construct index, and the semantic analysis and semantic extension of query request of user are accomplished by using index. The semantic level information retrieval is realized. (4) based on the semantic dictionary and relational category, the concept word hierarchy is constructed, which lays the foundation for semantic query expansion. A semantic vertical search service based on computer security domain ontology is designed and implemented in this paper. Experiments show that the design of semantic vertical search system in this paper can solve the defects of mechanical search and query accuracy in keyword retrieval. It shows that the semantic vertical search based on domain ontology is of practical significance.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP391.1

【相似文献】