图数据库对象级别关键词检索算法研究

发布时间：2018-03-25 08:29

本文选题：关系数据库　切入点：图数据库　出处：《大连海事大学》2013年硕士论文

【摘要】：关系数据库技术与信息检索技术的融合,在应用需求的推动下迅速发展。使用户既不需要懂得复杂的结构化查询语言,又不需要懂得底层的数据库模式,便可以像使用Web搜索引擎一样对数据库中的数据进行查询。对于关系数据库信息检索的策略,国内外专家学者提出了许多不同的观点。其中,既有元组级别的又有对象级别的。关系数据库中数据量的与日俱增,使得数据图的规模越来越大,信息检索的效率也越来越低。关系数据库信息检索领域面临大数据的挑战,己成为一个不可回避的事实。随着图数据库技术的不断成熟,其应用领域正在不断地扩大。与生俱来的灵活的图模型不但满足了社交类网站的应用需求,而且对图算法的适应能力也非常强。本文研究了图数据库技术和全文索引技术,分析了对象级别信息检索及其图数据库检索的研究现状,提出了一种由关系数据向图数据转换的数据抽取方式,并对现有的对象级别建模方式进行了改进,设计了一个嵌入图数据库的对象级别信息检索算法,相比元组级别的信息检索方式,对象级别的检索方式具有数据图规模小、结果完整性高和无重复结果等优点。该算法在考虑检索关键词的重要性的基础上,采用启发式的方式进行了规则查询,结合了图数据库与关系数据库,为海量数据条件下进行关系数据库信息检索提供了一种有效的解决方案,并拓展了图数据库的应用领域。为验证算法的有效性和原型系统的可用性,本文利用DBLP数据集对该算法的查询效果和查询效率进行了实验验证。论文采用P@k对检索效果进行了验证,并对检索效率进行了对比和分析。最终的实验结果表明,图数据库对象级别关键词检索算法具有良好的检索效果和较高的应用前景。
[Abstract]:With the combination of relational database technology and information retrieval technology, it develops rapidly under the impetus of application requirements, so that users do not need to understand the complex structured query language and the underlying database schema. We can query the data in the database just like using the Web search engine. For the strategy of information retrieval in relational database, experts and scholars at home and abroad have put forward many different viewpoints. With the increasing of data volume in relational database, the scale of data graph is becoming larger and larger, and the efficiency of information retrieval is becoming lower and lower. The field of relational database information retrieval is facing the challenge of big data. Has become an unavoidable fact. With the development of graph database technology, its application field is expanding continuously. The inherent flexible graph model not only meets the application needs of social networking sites, Moreover, the adaptability of graph algorithm is very strong. This paper studies graph database technology and full-text index technology, and analyzes the research status of object level information retrieval and graph database retrieval. In this paper, a data extraction method from relational data to graph data is proposed, and the existing object level modeling method is improved, and an object level information retrieval algorithm based on embedded graph database is designed. Compared with tuple level information retrieval, object level retrieval has the advantages of small scale of data graph, high result integrity and no repetition. This paper adopts heuristic method to query rules, combines graph database with relational database, provides an effective solution for relational database information retrieval under the condition of massive data, and extends the application field of graph database. In order to verify the validity of the algorithm and the usability of the prototype system, the query effect and efficiency of the algorithm are verified by using DBLP dataset. Finally, the experimental results show that the object level keyword retrieval algorithm of graph database has good retrieval effect and high application prospect.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP391.3;TP311.13

【参考文献】