The Research For Spatial Ontoloies Database In Cyber City

Cyber city is becoming ubiquitous with the development of Urban GIS. To start a cyber city project confronts many obstacles, including how to describe the rich detail urban environment. A close inspection of existent urban information shows such a fact that when the data is of apparently similar categories, we often use different semantics due to our different cognition of geography. This "Babel Tower" makes the association process and the establishment of an integrated cyber city project very problematic. On the other hand, it is desirable that sharing and re-using existing urban data from heterogeneous information systems to manage urban infrastructures. But the lack of formula methods to re-use knowledge and data makes this task difficult. Above all, the design of cyber city project is often challenged by opposing needs: (1) the individual need to maintain datum structure and contents strictly, to accommodate special situations; (2) the corporate need to retain compatibility among distinct field datum, to accommodate their interoperability.Therefore, this thesis selects spatial ontologies database as a research topic, and explores its key theoretical and practical issues. The investigations performed in this thesis and the relevant conclusions are outlined as fellows:The word "ontology" roots in Philosophy, where it means a science or study of being. Ontologies is emerging as an important tool for constructing sharable and re-usable knowledge repositories and supporting their interaction, the reason for which is that ontologies defines common expression nomenclatures that provide mutual understanding of an application domain among groups of users. The ontologies describes concepts and relationships that can be used by a group of information systems as a semantic basis on which they can communicate and exchange data. In this respect, this thesis investigates a multi-layered ontologies definition framework consisting of ontologies layers which are composed of a generic spatial functional structure and one or more domain ontologies for knowledge sharing & re-using.The urban environment is very complex. It has some natural occurrence-es, such as rivers, which not only belong to bona fide objects in appearance, but also belong to fiat objects. The classification of bona fide objects and fiat objects bases on boundary characteristics of geographic features. Bona fide objects have boundaries that exist independently of all human cognitive. Fiat boundaries, on the other hand, correspond to no genuine heterogeneity on the side of the geographic entities. They are abstract, created by acts of human decision, and are usually related to laws or political decrees. Based on this classification, this thesis proposes a heuristic methodology for creating and managing domain ontologies and integrating ontologies into database design.Knowledge sharing is essential for cyber city applications. This involves sharing data across collections of the heterogeneous information systems that are used to manage large urban infrastructures. The objective of this thesis is to define a spatial ontologies database framework to describe key features of urban applications which based on mereotopology & location theory, and providing a foundation for semantic reconciliation among heterogeneous urban information sources. The functional structure embodies general ontological concepts described as abstract data types. The domain ontologies is created by specializing in the properties and constraints of the functional structure. Inter-ontologies relationships are defined to integrate information across functional ontological layers and used to query multiple domain ontologies. The rich detail urban information is often stored in traditional relational databases. An aim of this thesis is to propose a strategy based on nesting -relation database model - a N1NF database model which reflects the nested structure of semi-structured spatio-temporal data - and map calculus suitable for spatio-temporal data to integrate ontologies into relational database. This model can not only store & maintain spatio-temporal data in ORDBMS efficiently, but also integrate different semantics based on adapting recursion operation of map calculus.At the end of this thesis, a cyber city project case is expatiated as a support.