The Earth Observation Metadata and Data Quality Drive
Session ID#: 282718
Session Description:
Metadata Strategy: Many user issues in Earth science data portals stem from underlying metadata flaws rather than functional bugs. To improve user experience, we will elevate metadata quality by enforcing strict schema validation, establishing a centralized source of truth for data catalogs, and leveraging AI/ML for continuous, automated enhancements. We will measure success by tracking the reduction in user-reported search and discovery issues.
Data Strategy: Currently, unstandardized datasets require expensive, custom processing pipelines. By standardizing incoming Earth science data, we can drastically cut ingestion, archiving, and distribution costs. We will achieve this through a unified evaluation process that checks datasets against community standards, tests data upon ingestion, provides automated fixes, and strictly enforces standardization for all new science products.
Index Terms:
1916 Data and information discovery [INFORMATICS]
1946 Metadata [INFORMATICS]
1950 Metadata: Quality [INFORMATICS]
1954 Natural language processing [INFORMATICS]