Geographic ontology is a very complex and intricate concept. As a philosophical concept, ontology has the most
ambiguous meanings. In the paper, firstly the concept of geographic ontology is explored and the three meanings of
geographic ontology are put forward: philosophy ontology, information ontology and spatial ontology. Then it introduces
some ontology representation languages and concludes by suggesting OWL as the appropriate language for building
geographic ontology. But geographic ontology is quite distinct from the general information ontology in that it should
represent spatial properties as well as attribute properties. Unfortunately, OWL cannot describe spatial properties and
spatial relations of geographic ontology because they are mainly for representing the general information ontology. It
puts forward to build some formal axioms about spatial properties of geographic ontology by use of three theoretical
tools: mereology, location theory and topology. With these axioms, we can formally represent geographic ontology with
spatial properties. Finally, it shows how we can use geographic ontology to describe map services and to improve the
quality of map services.
Web map is the hybrid of map and the World Wide Web (known as Web). It is usually created with WebGIS techniques.
With the rapid social development, web maps oriented the public are facing pressure that dissatisfy the increased
demanding. The geocoding database plays a key role in supporting query services effectively. The traditional geocoding
method is laborious and time-consuming. And there is much online spatial information, which would be the
supplementary information source for geocoding. Therefore, this paper discusses how to improve query services by web
mining. The improvement can be described from three facets: first, improving location query by discovering and
extracting address information from the Web to extend geocoding database. Second, enhancing the ability of optimum
path query of public traffic and buffer query by spatial analyzing and reasoning on the extended geocoding database.
Third, adjusting strategies of collecting data according to patterns discovered by web map query mining. Finally, this
paper presents the designing of the application system and experimental results.
With the far and wide applications of Location-Based Services (LBS), the call for more semantic and accurate services is
emerging. From a semantic viewpoint, the major characteristic of, and challenge for, LBS is the fact that they serve as
mediator between a possibly unknown user and possibly a priori unknown services. While some geographic information
technology standards provide the basis for syntactic interoperability, they do not yet provide methods for dealing with
problems of semantic heterogeneity. In this paper we design ontologies for LBS which are used for the identification and
association of semantically corresponding concepts to overcome the semantic problems. In order to better understand the
semantic content of the data in LBS, we analyze several elements both data and services involved. Then, we model these
data and services in a way that captures their peculiarities and allows their sharing between users and services and
exchange among different LBS, when desired. For this, we use the Protégé-OWL plug-in for creating hybrid hierarchy of
ontologies to enhance the semantic content both the user information and the services have. To argue about the design
choices and show their applicability, we present a simple example from a characteristic real world application.