基于领域本体的文化资源查询扩展方法及其应用

发布时间：2018-04-04 17:01

本文选题：领域本体　切入点：土家音乐文化资源　出处：《华中师范大学》2017年硕士论文

【摘要】：随着“互联网+”和信息化进程的不断加快,各种各样的信息呈“爆炸式”增长,如何在浩如烟海的信息资源中快而准的检索到用户所需的信息,成为目前大多数检索系统亟待解决的难题。基于传统的单一关键词匹配查询方法虽然在词汇匹配层面暂时解决了用户的查询需求,但是查询结果不尽人意,无法解决由于“一词多意”或者“一意多词”造成的查询率和召回率失衡的问题,难以满足用户的检索需求。本体作为在人工智能领域能与计算机交流的逻辑语言,在检索领域发挥了重大的作用,不仅能支持语义概念间的关系,而且能被共享和形式化。论文把本体技术引入到基于领域本体的文化资源查询系统中,充分挖掘了土家音乐文化资源之间的语义关联性,提高了系统的检索性能,从而在语义层面上满足了用户的查询需求,为土家音乐文化资源的学习、保护和发展打下了良好的基础。首先,论文对本体的相关理论概念进行了研究,梳理了本体的构建原则、构建方法、构建过程、编码语言和构建工具等,同时又根据土家音乐文化资源独有的资源属性,梳理土家音乐文化资源的相关概念、数据属性、对象属性等,并对其进行模型化,最终构建出了土家音乐文化资源领域本体。其次,论文对查询扩展技术进行了研究,提出将土家音乐文化资源领域本体引入到查询扩展中,该方法的基本思路是在基于局部分析的查询扩展方法中引入本体技术。本体技术能够提供可以共享的、形式化的、明确的、概念化的表示信息,使其以语义化的形式提供查询扩展库,能够改善基于局部分析的查询扩展过程,将二者结合能够有效的提高查询的准确率。通过计算本体内部概念之间的相似度,把在设定阈值之内的概念与检索词—并加入到查询词库,利用Jena对本体的解析技术,实现对土家音乐文化资源的本体库的检索功能,并将查询结果返回给用户。最后,与Solr搜索引擎技术结合,最终设计并实现了基于领域本体的文化资源检索系统。利用论文中提出的将本体技术引入到查询扩展的方法,实现对用户的查询词进行推理和扩展,最终得到查询结果。通过与基于关键词查询匹配方法对比,验证了该方法的有效性,通过对比实验结果可得,基于土家音乐文化资源领域本体的检索系统能有效的提升检索的性能和满足用户的查询需求。
[Abstract]:With the continuous acceleration of the "Internet" and the information process, various kinds of information are growing explosively. How to retrieve the information needed by users quickly and accurately in the vast information resources?It has become an urgent problem for most retrieval systems.Although the traditional single keyword matching query method solves the query needs of users temporarily at the lexical matching level, the query results are unsatisfactory.It is difficult to solve the problem of the imbalance of query rate and recall rate caused by "one word with multiple meanings" or "one meaning and many words", and it is difficult to meet the users' search demand.Ontology, as a logical language that can communicate with computer in the field of artificial intelligence, plays an important role in the field of retrieval. It not only supports the relationship between semantic concepts, but also can be shared and formalized.In this paper, the ontology technology is introduced into the query system of cultural resources based on domain ontology, which fully excavates the semantic correlation among the cultural resources of Tujia music, and improves the retrieval performance of the system.Therefore, it meets the query needs of users on the semantic level, and lays a good foundation for the study, protection and development of Tujia music culture resources.First of all, the paper studies the related theoretical concepts of ontology, combing the ontology construction principles, construction methods, construction process, coding language and construction tools, and at the same time according to the unique resource attributes of Tujia music culture resources.Combing the Tujia music cultural resources related concepts, data attributes, object attributes and so on, and model them, finally build Tujia music cultural resources domain ontology.Secondly, this paper studies query extension technology, and proposes to introduce Tujia music cultural resource domain ontology into query expansion. The basic idea of this method is to introduce ontology technology into query extension method based on local analysis.Ontology technology can provide shared, formalized, explicit and conceptualized representation information, which can provide query extension library in semantic form, and can improve the query extension process based on local analysis.The combination of the two can effectively improve the accuracy of the query.By calculating the similarity between the concepts within ontology, the concepts and search words within the set threshold are added to the query thesaurus, and the retrieval function of the ontology library of Tujia music and culture resources is realized by using the Jena technology of ontology analysis.The query result is returned to the user.Finally, with the combination of Solr search engine technology, a domain ontology based cultural resource retrieval system is designed and implemented.The method of introducing ontology technology into query extension is proposed in this paper to realize the inference and extension of user query words, and finally the query results are obtained.The validity of the method is verified by comparing it with the matching method based on keyword query.The retrieval system based on Tujia music culture resource domain ontology can effectively improve the performance of retrieval and meet the query needs of users.
【学位授予单位】：华中师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：G252.7

【参考文献】