【作者】 平湖；

【导师】 陈家训；

【作者基本信息】 东华大学 ， 控制理论与控制工程， 2007， 博士

【摘要】 异构数据库由一组互连的、自治的成员数据库组成。这些成员相互通信,以便交换信息和回答查询。一个成员数据库中的对象可能与其他成员数据库中的对象很不相同,使得很难将他们的语义吸收进一个整体的异种数据库中。因此我们需要考虑多种多样的语义,制定从一种表示到另一种表示的精确的转换规则,本文着重研究的是基于本体的电子政务公文异构数据源信息交换。即从已有系统中构建出了一个统一的、能够进行互操作的语义层,方便各政府机关实现跨平台的电子公文交换,指导和规范政府各部门内部办公网络系统的建设,逐步实现统一平台、统一网络、统一标准,加快政府各部门间的公文流转速度,提高办公效率和资源利用率。W3C制定的OWL DL是一种面向语义Web的知识表示标记语言,具有较强的知识表达能力并适合大规模应用的推理效率,是语义Web领域对适合语义Web应用的知识表示标记语言进行研究的最新综合,在知识表示领域,为了对知识表达能力和推理效率做合适的折衷,进行了大量关于描述逻辑的研究.本文详细分析了OWL DL和描述逻辑间的对应关系,从而可以将描述逻辑领域研究的大量成果应用到OWL DL上来,为进一步研究OWL DL的知识表示、推理等问题奠定了基础。本体是对一个特定领域中重要概念的共享的形式化的描述,是对某个领域中的概念的形式化的明确的表示,每个概念的特性描述了概念的各个方面及其约束的特征和属性,领域本体由属性、对象、关系和子领域本体组成.引入本体的思想,借助本体对领域知识进行详细描述,以抽象到概念化的语义层次,为进一步研究语义化的信息交互提供了基本的语义层次。政务数据元是政务信息交换的关键技术之一,它提供了电子政务互操作的一个基础,随着越来越多的数据元标准的出现,数据元的互操作问题已成为电子政务互操作的一个重要方面。分析数据元互操作产生的原因,本文从技术角度总结和研究了数据元互操作的主要解决途径,提出了一种新的实现这些数据元之间的互操作性解决方法,即基于政务本体解决政务数据元的语义互操作问题。概念聚类是领域数据挖掘和构建领域本体的关键,面向属性归纳是一种广泛使用的数据挖掘方法。结合面向属性归纳的特点,给出了一种改进的基于政务数据元的概念聚类方法。利用领域本体相互之间存在的语义相关性,从语义匹配的角度探讨了构造新领域本体的可能性.首先给出了领域本体的一种结构化定义,然后讨论了领域本体之间存在的语义相关性,并给出了领域本体语义相关度的概念.以此为基础,提出了一个利用OWL DL和描述逻辑间的对应关系、领域本体相互之间存在的语义相关性、通过杂交和合并不同领域的本体等构建政务领域本体的方法,在过去的几年中,本体技术在智能信息检索,电子商务,网上协作等信息技术领域的应用前景正越来越广。许多应用要求对相关信息源进行综合查询,因而解决这些异构数据源的语义互操作是必须的。本体映射是实现异构本体互操作的有效方法,提出了一种基于语义的本体映射方法,该方法考虑了本体中的概念属性,采用概念名称相似性、概念属性集合相似性等确定本体之间的语义映射关系.由于同一领域的不同信息源的本体创建者采用的语义不同,将它们集成涉及许多复杂因素。本文深入研讨了政务领域本体的互操作性,提出了一种创新的基于语义相似度、描述相似度、本体集成、本体推理相结合的局部本体和领域本体映射关系建立的方法,来取得政务领域本体语义的一致性。作为一种结果、从已有系统中构建出了一个统一的、能够进行互操作的语义层。最后作为一种应用,给出了基于本体的政务信息交换系统模型。更多还原

【Abstract】 Heterogeneous database is composed by a group of interconnect and autonomous member databases. These members communicate with each other, so as to exchange information and answer query. Since the objects in one member database may be very different from those in the others, it is difficult to absorb their semantics into one integrated database. Therefore, it is necessary for us to consider different semantics, and to make an accurate transfer rule from one expression to another. In this dissertation, we put our research emphasis on the heterogeneous data resource information exchange of electronic governmental documents based on ontology ,as to building a direct and standardize semantic level from now system and it may do semantic interoperation, Besides, we make a unified technical standard to guarantee the inter-platform electronic document exchange between the intranet of different government offices, direct and standardize the construction of government office network system, so as to realize the goal of unified platform, unified network and unified standard, to speed up the document flow between different government offices, to improve office efficiency and make full use of resources.OWL DL being established by W3C, is a knowledge representation markup language for the semantic web. OWL DL is a new synthesis of research on KRML for the semantic Web, which needs to be expressive enough for representing knowledge and computationally tractable for inference of large scale applications. In this paper,; the corresponding relation between OWL DL and description logic is analyzed, So the research conclusions of description logic can be applied to OWL DL , and the foundation of investigating KR and reasoning problems for OWL DL is established.An ontology is a common, shared and formal description of important concepts in an specific domain. An ontology is a formal explicit representation of concepts in a domain. properties of each concept describes characteristics and attributes of the concept known as slots and constrains on these slots. Domain ontologies is composed of attributes、object、relationships and sub-Domain ontologies. Domain knowledge is described in detail with ontology, and abstracted conceptualization semantic level, which provides semantic concepts for the subsequent research on semantic inter-operation.E-government data element is one of the key technologies in information exchange of E-government. it support a base for interoperability of E-government, With emerging of various kinds of data element the interoperability of data element, will be a focus, and important aspect of interoperability of E -government. The article pointed out the reasons of producing data element, interoperability, then summarized complete resolutions of data element interoperability, Pointed out that adopt a new technology and a way to research and carry out the data element interoperability ,this is a way using the domain ontologies of E -government to resolution of data dement semantic interoperability.Conceptual clustering is key about domain data mining and building domain ontologies Attribute-oriented induction is a method for data mining that has recently been widely applied, makes a study of that how the process of conceptual clustering incorporate the characteristics of the AOI process. An improved conceptual clustering method is also provided based on data element of E -government.Used existing the semantic relevance between domain ontologies, and proposes the possibility of building ontologies following the view of semantic match. Firstly, a structural definition of domain ontology is given. Secondly, the semantic relevance between domain ontologies is discussed and a definition of relevance degree is given Based on the evolution of population, using the relation between OWL DL and description logic、semantic relevant degree between domain ontologies、such as crossover and meaging domain ontologies from different domains at present etc, giving a new way to build the domain ontologies of E-government .In past few years ,Ontology is widely used in intelligent information retrieve, e-business, and network collaboration. Meanwhile, many applications require the integrated query from related information resources, which arises the problem of semantic interoperation in heterogeneous information resources is necessary. Ontology mapping is an effective method to realize the interoperation of heterogeneous Ontologies. A semantic based Ontology mapping method is proposed and concept properties are considered. Similarities of concept names, concept properties set are used to calculate semantic mapping relation between two Ontologies. Semantic radius makes the semantic mapping method more flexible.There are many complex factors in integrating them, because different ontology creators adopt different semantics from his individualview. The article discuss deeply the interoperation of the e-government domain Ontologies . A new effective method about building semantic mapping relation between part Ontologies and domain ontologies is proposed that adopt to combine with the semantic similarity, descriptive similarity, ontology integration, reasoning on ontology etc. to get ontologies semantic coherent.as a result .building a direct and standardize semantic level and it may do semantic interoperation from now system.Finally .as an application, given a model of E-government information exchange system based on ontology .更多还原