The Research On Geo-Information Services Based On Ontology

The paper leads the ontology thoughts and the methodology into the geographic information science domain, elucidates the exact meaning of geographpic ontology and the function and the sense of ontology in the geographic information science domain. The paper mainly discuses and deeply investigates the foundational theory problem of geographic information service based on ontology, we called it semantic difference problem, by combining instance and using concept lattice and formalization concept analysis method, and suggests a three-layer ontology framework by acting the basic concept design for geographic domain ontology; The paper thoroughly analyses ontology semantic relation which includes semantic similarity degree, structure similarity one, semantic correlation one etc, and establishes the foundation for resolving geographic information Web service matching and discovering; by using OWL-S which is commended by OGC and based on OWL language, the paper describes ontology Web service semantically and creates a service prototype based on ontology satisfied for landuse information presentation and query; Finally, the geographic information service composition problem which is based on ontology is discussed, and the ontology's research potential and value in geographic information science domain are showed elementarily.The main works of the paper is as follow:1,This paper thoroughly discusses the foundational theory problem of geographic information service based on ontology-semantic difference problem. Starting from an instance, we summarize the causation of semantic heterogeneity which is caused by cognition, the first is the differentia of classification detailed levels (the depth of classification tree), the second is different opinions on classification or exerted single inheriting relation; the third is different conceptualization. This paper analyses and compares thoroughly two representative geographic space cognition models, points out that geographic space can be divided into atomic space, geographic concept and application geographic concept, and establishes foundation for geographic information services description and composition based on ontology.The paper brings concept lattice and formalization concept analysis method to resolve semantic heterogeneity problem, the method helps to classification of complex concept and relation and can make semantic consistency. Formal concept analysis and concept lattice can express primely semantic relations and characteristics between concepts or classification, and can realize semantic consistent expression of geographic concept or classification.2,Taking landuse domain ontology as example, the paper conducts concept design of geographic domain ontology, suggests a three layers ontology framework, including the top-layer general ontology, middle-layer domain ontology and bottom-layer one, satisfies application research needs of land-use domain ontology. The paper discusses ontology modeling, ontology definition, semantic relation and construction methods of geographic domain ontology, describes in detail land-use domain ontology by UML. The expression language that relate to ontology, especially OWL language that relate to Web is also discussed in detail. The paper describes land-use type ontology by OWL and realizes it in Protege software.3,This paper resolves the problem of geographic information Web service matching and discovering based on geographic domain ontology and similarity degree, discusses semantic similarity degree, structure similarity degree, semantic correlation degree problems, establishes a foundation for finding and fulfilling the Web geographic information service that satisfied user's need. Starting from the concept of semantic similarity degree, the paper disusses the semantic similarity degree algorithm based on the distance among WordNet words, the semantic similarity degree algorithm based on HowNet sememe and the structure one, and also presents a similarity calculation instance by using the land-use domain ontology. Corresponding to semantic similarity degree, correlativity consists in between ontology and ontology. The semantic similarity degree and correlativity between ontology and ontology reflect all kinds of complicated relation between ontology concepts from different sides. Geographic concept system belonged to domain ontology is an authoritative description on the relation between concept and concept systems from domain experts. Correlativity reflects the correlation degree between ontologies, and Correlativity calculation mainly resolves the problem on how to build a rational mechanism to quantize the relationship seeming subjectively. The paper realizes the correlativity calculation using land-use domain ontology instance.4,Using Web Servise Ontology based on OWL language-OWL-S, this paper semantically describes Web service based on the ontology, including Web service profile, Web service modeling, Web Service groundingt etc., analyses with emphasis the process type, process ontology and service control structure. Ontology Concept involved in land-use change service and the basic service that is needed are defined and described in detail, and inputs, outputs and pre-setup conditions are set for every process. This paper also uses WSDL and OWL-S to describe and express semantic network services.5,The paper used the data of land-use and land-cover of Henan province from explanation of Landsat TM4,5,7 band standard false color image collected in 2000 year, discussed the display methods of classification and level of land-use data based on ontology, realized the query service of land-use information based on ontology, showed the capability and advantages that GIS displays geographic information of different semantic granularity (level) and provides geographic information services oriented-everybody. 6,It is very difficult that the single geographic information service meets the user's need, so how to composite the single Web service together to form a new Web services according to the application requirements is very important for bringing into play great potential of the Web Service. The paper mainly discusses the Web service matching problem, defines and describes the single direction's semantic matching relation in detail; Based on the single direction's semantic matching relation, the realization process of Web service composition algorithm are discussed deeply; combining a application instance and With the help of OWL-S, this paper describes every atomic Web service in detail, and educes the realization progress of the automatic combination algorithm. The geographic information Web service composition general framework is summarized based on automatic search, match, finding and execution progress of GIS Web service described by OWL-S.