Report

One of the main goals of the CEDAR project is to build an infrastructure for the creation and storage of machine-readable metadata templates. Toward that end, we have defined a JSON-based metadata model in which templates, elements, and fields are defined in JSON Schema, and the JSON-LD metadata instances are specified by the corresponding template. The specification allows CEDAR metadata to be used by CEDAR, and exchanged with other metadata systems.

This approach yields several advantages:

While good metadata is essential in finding, interpreting, and reusing data, the authoring of metadata is considered highly tedious and often incomplete. It is imperative to make the authoring of metadata a manageable task. To-wards easing the burden of authoring high quality metadata, we have developed a data-driven framework that learns as-sociation between data elements to suggest context-sensitive metadata values. We demonstrate our framework in the con-text of microarray annotations from the Gene Expression Omnibus (GEO).