APPROACHES TO IMPROVE SEMANTIC DESCRIPTION AND REASONING CAPABILITY IN THE DEEP TIME KNOWLEDGE BASE (Invited Presentation)

“Earth in Time” has been identified as a guiding theme for geoscience research in the next decade (NASEM, 2020). Deep time (i.e. geologic time) is an essential topic in the long history of the Earth and can be used as a common reference to connect various parameters across data facilities. However, the current cyberinfrastructure is short of formal representations for deep time concepts. With more and more geoscience data being made open on the Web, there is an urgent need for machine-readable deep time knowledge graphs to help reduce semantic heterogeneity among deep time concepts and facilitate data integration. Those knowledge graphs should be derived from community-level standards, should be based on common encoding formats on the Web, and should be easy to access and work together with existing geoscience open data. This presentation will report the latest progress of our research on a deep time knowledge base: 1) the coordination between global, regional and local geologic time scales in a common framework, 2) a version control structure for the evolving global geologic time scale, and 3) a list of functions to incorporate reasoning capability into the knowledge base. All the outputs are made open source and are accessible through the project website: http://www.deeptimekb.org.

Acknowledgment: This work is supported by the National Science Foundation (NSF OAC #1835717).

References: NASEM (National Academies of Sciences, Engineering, and Medicine), 2020. A Vision for NSF Earth Sciences 2020-2030: Earth in Time. The National Academies Press, Washington, DC, 172 pp. DOI: 10.17226/25761.