Sonet

Summary

SONET is a comprehensive semantic knowledge graph that is originally created to support the conflation of heterogenous geographic feature categories from diverse data sources in the PlanetSense project. SONET currently includes over 16,000 geographic feature categories with 113,020 knowledge links among them. It provides semantic interlinks through OpenStreetMap tags for over ten popular spatial data sources on the web that includes GEM Taxonomy, NAICS, among others. SONET is created by integrating ontology engineering, natural language processing, machine learning and semantic web technology, and it serves as a foundational information infrastructure for handling geographic data heterogeneity in various research and applications, including human dynamic modeling, locational intelligence, and temporally adaptive multi-scale global land use modeling.

Objectives

Support the conflation of heterogenous geographic feature categories from diverse spatial data sources
Foundational spatial information infrastructure that links diverse spatial data sources on the web
Provide a unified tagging schema for geographic feature categories from different data sources
Link different spatial data schemas through a shared semantic tagging schema
Seamless integration of heterogeneous spatial data with Web 3.0 linked open data cloud

Features

Semantic bridge for heterogeneous geographic data schemas

Interlink over ten popular spatial data schemas on the web such as Google Place, Facebook, Foursquare, GEM Taxonomy, and NAICS using a unified OSM tagging schema

Semantic translation between geographic feature categories and OSM tags

Design and develop unique OSM tags schema for geographic feature categories with different semantic granularities

OSM tags as semantic bridge for heterogeneous geographic feature categories

Map diverse geographic feature categories to a set of OSM tags based on translation schema

Multiple semantic granularities based on hierarchical organization of geographic feature categories

Three-level hierarchical organization of geographic feature categories that allows for grouping and analysis of features at different semantic granularities

Formal ontological model to support link of SONET with other Linked Open Data

Use formal OWL ontology language to model to hierarchical relationship among geographic feature categories, allowing future integration with Web 3.0 linked open data

Scientific artifacts

Year	Citation
2020	Thakur, G., Sparks, K., Berres, A., Tansakul, V., Chinthavali, S., Whitehead, M., Schmidt, E., Xu, H., Fan, J., Spears, D., & Cranfill, E. (2020). COVID-19 Joint Pandemic Modeling and Analysis Platform. Proceedings of the 1st ACM SIGSPATIAL International Workshop on Modeling and Understanding the Spread of COVID-19, COVID-19 2020, 43–52. https://doi.org/10.1145/3423459.3430760
2019	Palumbo, R., Thompson, L., & Thakur, G. (2019). SONET: A Semantic Ontological Network Graph for Managing Points of Interest Data Heterogeneity. Proceedings of the 3rd ACM SIGSPATIAL International Workshop on Geospatial Humanities - GeoHumanities ’19. https://doi.org/10.1145/3356991.3365474
2023	Fan, J. , Bentley, J. & Thakur, G. (2023). SONET++: A Knowledge Graph of Geographic Categories based on OSM Tag Representation. GIScience Conference 2023.

Contact Us

Junchuan Fan

Location Intelligence Group
Human Dynamics Section,
Geospatial Science and Human Security Division
National Security and Sciences Directorate
Oak Ridge National Laboratory

[email protected]

Gautam Thakur

Location Intelligence Group
Human Dynamics Section,
Geospatial Science and Human Security Division
National Security and Sciences Directorate
Oak Ridge National Laboratory

[email protected]