The Construction Of Land Cover Knowledge Graph

Land cover refers to the synthesis of various material types and their natural attributes and characteristics on the earth's surface,which reflects their distribution and evolution process of material types.It is an important basic data for global natural resources survey and ecological environment research.However,simply providing data and information services can not fully play its real role.Providing multi-granularity,multi-scale,and multi-dimensional land cover knowledge services is the current development trend.How to represent land cover knowledge,organize and associate these knowledge is the primary problem to be solved in the service of land cover knowledge.Based on this,this thesis introduces the knowledge graph into the research of land cover knowledge service,takes surface water as an example,studies the expression of land cover knowledge,effectively organizes and expresses all kinds of land cover data,and constructs the land cover knowledge graph.The main research contents of this thesis are as follows:(1)Knowledge modeling and formal expression of land cover data.Firstly,the characteristics of knowledge in the field of land cover are analyzed,and the knowledge classification system is studied according to the characteristics of knowledge points;secondly,build the land cover knowledge graph model layer based on ontology theory,design domain concepts,entities,relationships,attributes,etc.;finally,taking the water system as an example,according to the design of the knowledge graph pattern layer,the knowledge expression method is studied,and the RDF(Resource Description Framework)triplet model is used to formally express the knowledge,and it is stored in the graph model database to complete the construction of the knowledge graph data layer.(2)Multi-level correlation of land cover knowledge.Based on the spatiotemporal and semantic characteristics of land cover knowledge,a spatial relationship calculation model,a semantic reasoning model,and a spatiotemporal change expression model are proposed.Among them,the spatial relationship calculation model carries out association calculations on the topological,azimuth,and distance relationships between knowledge points;The semantic reasoning model uses description logic and rule reasoning to infer and associate the semantics between knowledge points.The spatiotemporal change expression model associates the spectral change and morphological change indexes with the knowledge points of land cover,so as to support in-depth exploration of the spatiotemporal change characteristics of land cover.(3)Design and implementation of prototype system for constructing land cover knowledge graph.Based on the graph model database for persistent storage of land cover knowledge,Jena is used to implement semantic inference,and a prototype system is developed with VUE,Java,and Mapbox.The system realizes three functional modules:knowledge storage of land cover,knowledge association and knowledge query.Examples have proved that the system is reliable in operation,has strong practical value,and can effectively support the construction of the knowledge graph of land cover.