物联网环境下面向服务的实体语义标注
发布时间：2018-08-16 18:17
【摘要】：随着物联网实体设备的快速增加,物联网开始面临异构化实体协同工作的问题。目前有研究将面向服务的跨平台思想和语义Web技术引入到物联网中,将物理世界中的实体功能用服务化的形式进行描述和访问。并通过服务发现技术发现可协作的服务链,进一步获取可执行服务的实体,从而使异构实体能够通过Web服务进行协作,解决实体协作问题。实体语义标注是运用语义Web服务发现技术的基础,是解决实体功能服务化和语义化描述问题的关键。因此,如何进行面向服务的实体语义标注成为亟待解决的问题。然而,现有的实体语义标注框架未考虑到物联网实体的状态可变性,实体与服务的耦合度过高。针对现有研究结果的不足,本文提出了一种面向服务的实体语义标注框架。该框架将实体不变的功能描述和可变的状态描述分离开来,使得物理层实体和逻辑层服务的关联松耦合。针对所提出的实体语义标注框架,提出实体语义标注本体模型。采用OWL描述逻辑对实体的基本信息、状态以及功能信息进行描述,并建立实体功能与语义Web服务之间的关联关系。为服务发现提供了发现可用执行实体的描述基础。针对实体功能与服务关联所可能导致的服务繁杂冗余的问题,提出了一种面向服务的聚类标注算法。首先针对现有方法的不足,提出了基于I/O混合匹配与关键字匹配的服务相似度计算方法。然后,对相似度高的服务进行聚类和中心服务标注。并通过实验与对比分析,证明了该算法对服务发现效率的提高。基于上述研究成果,本文设计并实现了基于实体语义标注的服务发现系统。首先,阐述该系统的总体设计;然后对语义标注实例生成模块、面向服务的聚类标注模块以及标注实例解析与筛选模块的设计及实现进行了详细阐述。最后,以可被感知的物流运输为背景,对服务发现系统进行测试与分析。首先,构建实体语义标注实例和语义Web服务。然后,设计服务请求,包括服务的功能请求以及对实体状态的要求。最后,对系统的运行结果进行分析。实验结果证明了本文所提出的实体语义标注框架的有效性。
[Abstract]:With the rapid increase of physical devices in the Internet of things, the Internet of things begins to face the problem of isomerization entities working together. At present, some researches have introduced service-oriented cross-platform thinking and semantic Web technology into the Internet of things, describing and accessing entity functions in the physical world in the form of service. The cooperative service chain can be found by service discovery technology, and the entity of executable service can be obtained further, so that the heterogeneous entity can collaborate through Web services and solve the problem of entity collaboration. Entity semantic annotation is the basis of using semantic Web service discovery technology and the key to solve the problem of entity function service and semantic description. Therefore, how to carry on the service-oriented entity semantic annotation becomes an urgent problem to be solved. However, the existing entity semantic annotation framework does not take into account the state variability of the entities in the Internet of things, and the coupling degree between entities and services is too high. This paper proposes a service-oriented semantic annotation framework for entities in view of the shortcomings of existing research results. The framework separates the entity invariant function description from the variable state description, which makes the physical layer entity and the logical layer service association loosely coupled. Aiming at the proposed framework of entity semantic annotation, an ontology model of entity semantic annotation is proposed. The basic information, state and function information of entities are described by OWL description logic, and the relationship between entity functions and semantic Web services is established. Provides a description basis for discovery of available execution entities for service discovery. In order to solve the problem of service redundancy caused by entity function and service association, a service-oriented clustering algorithm is proposed. Firstly, a service similarity calculation method based on I / O hybrid matching and keyword matching is proposed to overcome the shortcomings of the existing methods. Then, the services with high similarity are clustered and labeled with central services. Experiments and comparative analysis show that the algorithm can improve the efficiency of service discovery. Based on the above research results, this paper designs and implements a service discovery system based on entity semantic annotation. First, the overall design of the system is described, and then the design and implementation of the semantic annotation case generation module, the service-oriented clustering annotation module and the annotation case analysis and screening module are described in detail. Finally, the service discovery system is tested and analyzed in the context of perceived logistics transportation. Firstly, the entity semantic annotation instance and semantic Web service are constructed. Then, the service request is designed, including the function request of the service and the requirement of entity state. Finally, the operating results of the system are analyzed. The experimental results show the effectiveness of the proposed framework.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP391.1;TP391.44;TN929.5
本文编号：2186786
[Abstract]:With the rapid increase of physical devices in the Internet of things, the Internet of things begins to face the problem of isomerization entities working together. At present, some researches have introduced service-oriented cross-platform thinking and semantic Web technology into the Internet of things, describing and accessing entity functions in the physical world in the form of service. The cooperative service chain can be found by service discovery technology, and the entity of executable service can be obtained further, so that the heterogeneous entity can collaborate through Web services and solve the problem of entity collaboration. Entity semantic annotation is the basis of using semantic Web service discovery technology and the key to solve the problem of entity function service and semantic description. Therefore, how to carry on the service-oriented entity semantic annotation becomes an urgent problem to be solved. However, the existing entity semantic annotation framework does not take into account the state variability of the entities in the Internet of things, and the coupling degree between entities and services is too high. This paper proposes a service-oriented semantic annotation framework for entities in view of the shortcomings of existing research results. The framework separates the entity invariant function description from the variable state description, which makes the physical layer entity and the logical layer service association loosely coupled. Aiming at the proposed framework of entity semantic annotation, an ontology model of entity semantic annotation is proposed. The basic information, state and function information of entities are described by OWL description logic, and the relationship between entity functions and semantic Web services is established. Provides a description basis for discovery of available execution entities for service discovery. In order to solve the problem of service redundancy caused by entity function and service association, a service-oriented clustering algorithm is proposed. Firstly, a service similarity calculation method based on I / O hybrid matching and keyword matching is proposed to overcome the shortcomings of the existing methods. Then, the services with high similarity are clustered and labeled with central services. Experiments and comparative analysis show that the algorithm can improve the efficiency of service discovery. Based on the above research results, this paper designs and implements a service discovery system based on entity semantic annotation. First, the overall design of the system is described, and then the design and implementation of the semantic annotation case generation module, the service-oriented clustering annotation module and the annotation case analysis and screening module are described in detail. Finally, the service discovery system is tested and analyzed in the context of perceived logistics transportation. Firstly, the entity semantic annotation instance and semantic Web service are constructed. Then, the service request is designed, including the function request of the service and the requirement of entity state. Finally, the operating results of the system are analyzed. The experimental results show the effectiveness of the proposed framework.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP391.1;TP391.44;TN929.5
【参考文献】
相关期刊论文 前2条
1 袁凌云;王兴超;;语义技术在物联网中的应用研究综述[J];计算机科学;2014年S1期
2 毛峻岭;贾雪琴;刘红旗;;物联网语义架构和语义关键技术研究[J];信息通信技术;2014年05期
，本文编号：2186786
本文链接：https://www.wllwen.com/guanlilunwen/wuliuguanlilunwen/2186786.html
