基于领域本体的分布式推理与查询方法研究

发布时间：2018-01-18 13:31

本文关键词：基于领域本体的分布式推理与查询方法研究　出处：《中国民航大学》2017年硕士论文　论文类型：学位论文

【摘要】：民航突发事件应急管理领域本体是对突发事件应急处置中的概念、概念间关系和实例的描述,为民航应急管理领域知识的表达与共享、应急救援方案的推理奠定了良好的基础。随着信息获取和传播技术的飞速发展,网络多媒体资源迅速猛增,以SWRL规则为基础的领域本体推理难以适应大数据环境下分布式推理与查询的应用需求。为此,本文将MapReduce分布式计算引入民航突发事件应急管理领域本体的推理与查询,进行了以下的研究:(1)提出了一种基于RDFS的领域本体分布式推理方法,通过领域本体数据的预处理和RDFS推理规则设计,给出了将RDFS规则推理特征与MapReduce分布式计算相结合的推理过程,实验结果与WebPIE推理方法进行了比较,验证了该推理方法的有效性,解决了民航突发事件应急管理领域本体推理效率低的问题。(2)在对领域本体分布式推理研究的基础之上,进一步研究了领域本体分布式查询方法,详细论述了分布式查询关键算法的设计与实现过程,并采用贪心选择策略对查询过程进行优化,实现了MapReduce的一次迭代查询,实验结果验证了该方法在查询效率方面的明显优势,为基于分布式推理的民航突发事件关联分析提供了方法支撑。
[Abstract]:The domain ontology of civil aviation emergency management is the concept of emergency management, the relationship between concepts and the description of examples, which is the expression and sharing of knowledge in civil aviation emergency management. With the rapid development of information acquisition and communication technology, the multimedia resources of network are rapidly increasing. Domain Ontology reasoning based on SWRL rules is difficult to adapt to the application of distributed reasoning and query in big data environment. In this paper, MapReduce distributed computing is introduced into the reasoning and query of civil aviation emergency management domain ontology. The following research is done: 1) A distributed reasoning method of domain ontology based on RDFS is proposed, which is based on the preprocessing of domain ontology data and the design of RDFS inference rules. The reasoning process combining RDFS rule reasoning features with MapReduce distributed computing is presented. The experimental results are compared with WebPIE reasoning methods. The validity of the reasoning method is verified, and the problem of low efficiency of ontology reasoning in the field of civil aviation emergency management is solved, which is based on the research of distributed reasoning of domain ontology. Furthermore, the distributed query method of domain ontology is studied, the design and implementation of the key algorithm of distributed query are discussed in detail, and the greedy selection strategy is used to optimize the query process. An iterative query based on MapReduce is implemented. The experimental results show that this method has obvious advantages in query efficiency and provides a method support for civil aviation emergency association analysis based on distributed reasoning.
【学位授予单位】：中国民航大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.13

【参考文献】