The Construction Of Ontology-based Earth Science Knowledge Graph

Massive Earth science data has been collected and integrated from various means,which has brought the reform of the research paradigm for Earth science.However,due to inconsistent description standards of data,unclear data sharing mechanism and significant semantic heterogeneity,there still exist difficulties in data integration,sharing and reuse,which hinders the application of big data in geoscience research.The knowledge graph has brought new ideas and methods to Earth science research.The knowledge graph can be used to explicitly represent concepts and their complex semantic relationships,and is machine-understandable.Therefore,it has become a key technology for semantic translation,data integration and data reuse.In this paper,we analyzed the knowledge graphs and Earth science big data,and discussed the important roles of knowledge graph in Earth science big data research.After that,we proposed an ontology-based hierarchical framework to construct the Earth science knowledge graph.In this framework,the Earth science knowledge graph is divided into the subject layer,the core layer,and the extension layer.The igneous petrology was taken as an example,SKOS,RDF and OWL were used to model and represent this hierarchical framework.We also implemented the semantic translation,data finding and integration by knowledge reasoning based on the analysis of constraint relationships in the knowledge graph to solve the problem of semantic heterogeneity among multisource datasets.Based on that,we analyzed the demands of Earth science knowledge graph construction tool.The logic structure to storage,represent and manage of the Earth science knowledge graph were analyzed and the function were described,and the framework of the igneous petrology knowledge graph was constructed by the construction tool,hoping to facilitate the construction and application of Earth science knowledge graph.