Abstract: As the most effective way of knowledge organization and service at present, knowledge graph has become the cornerstone of artificial intelligence and has been widely used in semantic search, machine translation, information recommendation and so on. In the era of big data, there is an urgent need of Geoscience knowledge graphs for integrating, mining and analysis of scattered, multi-source and heterogeneous Geo-data and its unknown knowledge intelligent discovery. To promote the construction and application of Geoscience knowledge graphs, the Deep-time Digital Earth (DDE) International Big Science Program has taken knowledge graph as its one of the core research contents since its launch in 2019. After more than three years of construction, DDE has built a large number of Geoscience knowledge graphs, and it strong needs one-stop sharing and service system of these knowledge graphs. Firstly, this paper introduces the content framework, composition as well as characteristics of the DDE knowledge graphs. On this basis, the design of the one-stop sharing and service system for the Geoscience knowledge graph is designed that include the design of the system functional and technical architecture. Finally, the development and operation environment & tools, and key technologies of the system are discussed in detail. The practice has proved that the system can effectively realize the one-stop sharing and open access of DDE knowledge graphs. Meanwhile, it sets an example for other fields or domains’ knowledge graph integrating and sharing systems.

Key words: geoscience, knowledge graph, one-stop, integrated sharing, service system