基于本体论的生物信息学领域资源语义检索研究
发布时间：2019-03-02 18:38
【摘要】：随着生物信息学领域资源信息的不断增加，用户使用通用引擎搜索专业信息时出现很多弊端，例如：返回信息过多、存在无用的信息、准确率低等。针对这些情况，为满足用户需求的某一专业学科领域信息的专业语义搜索引擎应运而生。 语义检索是信息检索与人工智能技术、自然语言技术相结合的一种检索方法，本文将本体与数据挖掘、生物信息学术语的标准化、生物信息数据的整合都联系起来，最后形成一套完善的生物信息学本体论，用以整合所有与基因或蛋白质相关的数据，并且能够描述各生物体概念的属性及相互关系。 本文分析了国内外对信息检索的研究情况，指出传统信息检索的不足，提出能够实现语义层次检索的语义检索。接着对本体的相关概念及常用的关于本体的构建工具及构建方法做了详细的介绍，分析各自的优点与不足，综合考虑各种构建方法的优势，，并结合生物信息学的实际情况，提出基于《生物信息学分类》构建生物信息学领域本体。研究基于生物信息学本体的语义检索在生物信息学领域资源中的一些关键技术，其中着重描述了生物信息学本体结构中的概念在语义层次的查询扩展。随后，对语义相似度的计算方法进行了改进，分析语义关系对语义相似度的影响；并从概念及概念和概念之间的语义关系等方面对关键词进行了语义扩展。将以上所研究的这些技术运用于生物信息学数据平台，建立生物信息学系统领域的本体知识库，实现语义检索功能，通过测试验证该方法可以提高数据检索的查全率与查准率。
[Abstract]:With the increasing of resource information in the field of bioinformatics, there are many drawbacks when users search for professional information using general engine, for example, too much information is returned, there is useless information, and the accuracy is low. In order to meet the needs of users, a professional semantic search engine is developed to meet the needs of users. Semantic retrieval is a retrieval method which combines information retrieval with artificial intelligence and natural language technology. In this paper, ontology and data mining, standardization of bioinformatics terms and integration of biological information data are linked. Finally, a complete set of bioinformatics ontology is formed to integrate all the data related to genes or proteins, and to describe the attributes and relationships of various biological concepts. This paper analyzes the research situation of information retrieval at home and abroad, points out the deficiency of traditional information retrieval, and puts forward the semantic retrieval which can realize semantic level retrieval. Then the related concepts of ontology and the tools and methods of ontology construction are introduced in detail, the advantages and disadvantages of each method are analyzed, the advantages of various construction methods are comprehensively considered, and combined with the actual situation of bioinformatics, A domain ontology of bioinformatics based on taxonomy of bioinformatics is proposed. This paper studies some key technologies of semantic retrieval based on bioinformatics ontology in bioinformatics resources, in which the query extension of concepts in ontology structure of bioinformatics at the semantic level is described emphatically. Then, the calculation method of semantic similarity is improved, the influence of semantic relation on semantic similarity is analyzed, and the semantic extension of keywords is carried out from the aspects of concept and semantic relation between concepts. These techniques are applied to bioinformatics data platform to establish ontology knowledge base of bioinformatics system and realize semantic retrieval function. The method can improve the recall and precision of data retrieval by testing.
【学位授予单位】：中北大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP391.3
本文编号：2433350
[Abstract]:With the increasing of resource information in the field of bioinformatics, there are many drawbacks when users search for professional information using general engine, for example, too much information is returned, there is useless information, and the accuracy is low. In order to meet the needs of users, a professional semantic search engine is developed to meet the needs of users. Semantic retrieval is a retrieval method which combines information retrieval with artificial intelligence and natural language technology. In this paper, ontology and data mining, standardization of bioinformatics terms and integration of biological information data are linked. Finally, a complete set of bioinformatics ontology is formed to integrate all the data related to genes or proteins, and to describe the attributes and relationships of various biological concepts. This paper analyzes the research situation of information retrieval at home and abroad, points out the deficiency of traditional information retrieval, and puts forward the semantic retrieval which can realize semantic level retrieval. Then the related concepts of ontology and the tools and methods of ontology construction are introduced in detail, the advantages and disadvantages of each method are analyzed, the advantages of various construction methods are comprehensively considered, and combined with the actual situation of bioinformatics, A domain ontology of bioinformatics based on taxonomy of bioinformatics is proposed. This paper studies some key technologies of semantic retrieval based on bioinformatics ontology in bioinformatics resources, in which the query extension of concepts in ontology structure of bioinformatics at the semantic level is described emphatically. Then, the calculation method of semantic similarity is improved, the influence of semantic relation on semantic similarity is analyzed, and the semantic extension of keywords is carried out from the aspects of concept and semantic relation between concepts. These techniques are applied to bioinformatics data platform to establish ontology knowledge base of bioinformatics system and realize semantic retrieval function. The method can improve the recall and precision of data retrieval by testing.
【学位授予单位】：中北大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP391.3
【参考文献】
相关期刊论文 前7条
1 邓志鸿,唐世渭,张铭,杨冬青,陈捷;Ontology研究综述[J];北京大学学报(自然科学版);2002年05期
2 武成岗,焦文品,田启家,史忠植;基于本体论和多主体的信息检索服务器[J];计算机研究与发展;2001年06期
3 黄名选;严小卫;张师超;;查询扩展技术进展与展望[J];计算机应用与软件;2007年11期
4 宋峻峰,张维明,肖卫东,唐九阳;基于本体的信息检索模型研究[J];南京大学学报(自然科学版);2005年02期
5 王星灿;;略谈叙词法[J];情报科学;1986年06期
6 林冉;专题型搜索引擎调查分析[J];情报杂志;2003年09期
7 徐国虎;许芳;;本体构建工具的分析与比较[J];图书情报工作;2006年01期
相关博士学位论文 前1条
1 余传明;基于本体的语义信息系统研究[D];武汉大学;2005年
相关硕士学位论文 前2条
1 张功杰;基于本体的领域资源语义检索研究[D];暨南大学;2007年
2 毛平;基于领域本体的文本信息语义检索研究[D];南京理工大学;2007年
本文编号：2433350
本文链接：https://www.wllwen.com/kejilunwen/sousuoyinqinglunwen/2433350.html
