Research On Spatiotemporal Data Model And Query Method Based On RDF

There are a large number of data related to spatio-temporal characteristics in the field of geographic information system.These data are called spatiotemporal data,and their characteristics change with the change of time and space.Spatiotemporal data usually have complex data types and various forms of expression,which not only update and change rapidly,but also have rich spatiotemporal semantics.The function of spatiotemporal database is to manage spatiotemporal data,and its key lies in modeling and querying spatiotemporal data.Therefore,the construction of spatiotemporal data model and the query mechanism based on it,which is suitable for the dynamic change of spatiotemporal data and preserving semantic information requirements,is becoming a research hotspot in the field of spatiotemporal data management.With the wide use of Web 2.0technology,large-scale RDF(Resource Description Framework)data appears.RDF is a semantic framework proposed by W3 C,which is considered as the standard model of Web data resource representation and exchange.It has the characteristics of flexible application,good extensibility and strong inclusiveness.Therefore,the research of spatiotemporal data model and query method based on RDF is bound to break through the bottleneck of spatiotemporal data scale,fast dynamic update and rich spatiotemporal semantics in traditional databases,thus bringing an opportunity for efficient management of spatiotemporal data.Existing researches on RDF data management mainly focus on RDF data management,temporal RDF data management and spatial RDF data management,but the existing efforts are still difficult to meet the needs of spatiotemporal RDF data management.To this end,this dissertation studies spatiotemporal data model and query method based on RDF,and obtains the following innovative contributions.(1)The management ability of spatiotemporal data is mainly affected by the spatiotemporal data model,and the foundation of establishing spatiotemporal data model is the spatiotemporal data conceptual model.In Semantic Web,OWL(Web Ontology Language)has strong semantic representation and description ability,and is one of the important means to study the conceptual model of spatiotemporal data.Therefore,in the conceptual model of spatiotemporal data,starting with the formal representation of spatiotemporal data based on OWL,the axiom classes of spatiotemporal OWL are studied,including spatiotemporal equivalence class relations and spatiotemporal subclass relations.On this basis,the spatiotemporal OWL logic classes are proposed,including spatiotemporal disjoint classes,spatiotemporal intersection classes and spatiotemporal union classes.The conceptual model of spatiotemporal data provides a theoretical basis for the study of spatiotemporal data model.(2)In the aspect of spatiotemporal data model,firstly,a spatiotemporal data model based on RDF is proposed,including the representation method of st RDFS data model,spatiotemporal RDF data graph,spatiotemporal RDF query graph and so on.Secondly,four structures of spatiotemporal RDF data graph are defined,including linear structure,tree structure,star structure and loop structure.On this basis,the main classes in st RDFS are studied and described,including spatial classes and descriptions in spatiotemporal domain,time classes and descriptions in spatiotemporal domain and spatiotemporal classes and descriptions spatiotemporal domain.Thirdly,five types of st RDFS graph algebras are proposed,including union operation,intersection operation,difference operation,Cartesian product operation and filter operation.Then,an example is given to verify the correctness of the proposed model.Finally,the topological relations of spatiotemporal RDF data model are studied.The practicability of the proposed topological relations determination method of spatiotemporal RDF data model is verified by Gephi.(3)In terms of spatiotemporal data query methods,the matching methods of time intervals in spatiotemporal RDF data are studied through temporal intersection operation,temporal union operation and temporal span,and the matching methods of spatial intervals in spatiotemporal RDF data are studied through spatial intersection operation,spatial union operation and spatial span.Then,the query candidate fields are defined,and the calculation method of matching order is proposed.On this basis,the subgraph isomorphism in spatiotemporal RDF data is defined,and the query method of spatiotemporal RDF data is proposed based on subgraph isomorphism.Finally,the efficiency and effectiveness of the proposed query method are performed and illustrated by experimental evaluations and analysis on YAGO datasets.The above contributions aim to form a complete framework of spatiotemporal data from modeling to query based on RDF and break through the key technologies,which will provide theoretical and technical solutions for spatiotemporal RDF data model and query,thus promoting the further development of spatiotemporal data model and query research.The dissertation includes 85 figures,6 tables and 93 references.