Improving Discoverability of Open Government Data with Rich Metadata Descriptions Using Semantic Government Vocabulary

Abstract

The descriptive metadata gathered by open data catalogs are often simple key-value pairs that describe provenance information, but not concepts from the domain of the described dataset. Search engines relying on such metadata cannot make use of semantic connections among datasets. In this paper, we present a Semantic Government Vocabulary that is used for creating rich annotations of Open Government Data, allowing to find their mutual interconnections, as well as document their meaning in the machine readable form. We discuss how the Semantic Government Vocabulary is layered based on the different ontological types of terms occurring in the Open Government Data. Next, we show how the vocabularies can be used to annotate Open Government Data on different levels of detail and how to formalize the whole stack in the Web Ontology Language. We evaluate feasibility and usability of our approach using a study in the elections domain.

Keywords: Linked Data, Open Data, Ontology, Data Catalog, Dataset Discovery