Datastar – A Semantic Registry for Research Datasets

Datastar is an open-source, semantic-web-based platform supporting the description, discovery, access, curation, and reuse of research datasets. Datastar extends the ability of the VIVO researcher profiling system to represent relationships among researchers, grants, and publications to capture the scholarly context around research datasets and to highlight dataset citations. Datastar can be run either as a standalone dataset registry or as an extension to an institutional instance of VIVO. The Datastar project was initially funded in 2007 by NSF, and more recently in 2011 by the U.S. Institute of Museum and Library Services. In this presentation, we will report on the more recent work, which has involved both working with researchers to create a set of Data Curation Profiles to understand researchers’ needs and preferences with respect to documentation, sharing, dissemination, and reuse of datasets; and applying those profiles to the development of the Datastar application. Datastar promotes open access and easy discoverability of research datasets; it promotes reproducible research through interlinking research datasets with the full research context of researchers, publications, and grants; and it complements the capabilities of open institutional and disciplinary repositories as part of the overall research ecosystem.