Research On The Representation And The Construction Of Marine Environment Data Ontology Based On Spatial Relations

After years of development,China's marine information infrastructure construction has begun to take shape.The methods and scope of comprehensive survey of marine resources have been continuously expanded.The capacity of data acquisition and data volume have been greatly improved.The marine environment data mainly includes ocean hydrological data,marine meteorological data,marine physical data,marine chemical data,marine biological data,marine geological data,suspension data,marine geophysical data and etc.Due to different collection equipment and acquisition methods,standards and formats of those data are not uniform.The compatibility of these data is poor because they are often heterogeneous due to the different concerns of different organizations.These factors cause difficulties in data integration and sharing.Therefore,how to maximize the value of marine environmental data and improve the efficiency of integration has become a hot topic in academic circles.Ontology is an effective way to solve data integration and sharing,because it can clearly define the relationship between objects and concepts and enhance the correlation of data.Based on the background mentioned above,this paper devotes to studying the spatial relationship ontology construction for marine environmental data,by taking the diversity,huge volumes and heterogeneity of marine environment data as the research object and using the unified data and efficient query as targets.The main research contents and achievements are as follows:Firstly,by combing and studying the existing ontology construction methods,marine ontology and spatial relationship models,this study finds the following issues exist in the study of the current marine ontology construction:(1)The marine environment data is typical spatial data.In the existing expression model of spatial orientation,due to the influence of the size and distance of the target range,no accurate results can be obtained when discriminating the orientation relationship of the intersection region.It will cause deviations in ontology-based spatial relationship query results.(2)The existing marine ontology studies mainly focus on the ontology formal expression mechanism.They lack the description of the spatial characteristics and spatial relationship of the marine environment data.It leads to the poor practicality of the marine ontology.(3)The existing ontology construction methods and ontology model frameworks are mostly directed to their respective fields.At present,there is no ontology library specifically for the expression of marine environment data and spatial relationships.It makes the constructed ontology lack a strategy for expressing spatial location information of marine environment data and spatial relationships.Secondly,aiming at the strong spatial-temporal correlation characteristics of marine environmental data,this study sorts out and supplements the spatial relationship between data and proposes a conical model with a combination of internal and external azimuth relations.In view of the unique spatial characteristics of marine environmental data,combined with the improved conic model azimuth relation representation method,a method for constructing marine environmental data ontology considering spatial relationships was proposed and a framework of marine environmental data model was constructed.Taking the polar ocean environment data as an example,this paper validates the ontology expression and construction method of marine environment data which considering the spatial relationship proposed in this paper and constructs the polar ocean environment data ontology library and the polar marine environment monitoring system.The main content includes the following three aspects:(1)A representation model of the azimuth relationship between data is proposed by studying the spatial characteristics of marine environmental data.The model can reasonably express the orientation relationship between two spatial objects and solves the problem that the cone model and the minimum circumscribed rectangle model cannot express certain special azimuth relationships and achieves the accurate expression of the spatial relationship between spatial point,line,and surface,and provides a reliable method for determining azimuth relation based on ontology-based spatial association query.(2)On the basis of the concept of the marine environment data ontologies,the construction criteria and other related theories,a typical ontology construction method was analyzed in this study.Based on the characteristics of the marine environment data,combined with the improvement of representation method of the conic model orientation relationship,the ontology construction system was improved,and a method for constructing the marine environment data ontology considering the spatial relationship was proposed.(3)Based on the ontology construction method,the marine environment data catalogue,marine metadata,and the temporal and spatial characteristics of the ocean,this study focuses on the formal expression mechanism of the marine environmental data ontology and puts forward the framework of the marine environmental ontology knowledge model.In this framework,the spatial characteristics of the marine environment data are highlighted,and at the same time,the detailed description of the functions and structures of each module of the model framework is given.Finally,on the basis of the ontology database of marine environment data that takes spatial relationships into account,a polar marine environmental monitoring system platform has been established in this study.The construction of the ontology library enhances the semantic relationship and spatial correlation between polar marine environmental data.In the data query module,the recall rate and precision rate of data have been effectively improved.It provides a feasible solution for effectively solving the problem of data integration and heterogeneity in the marine environment.