DOE Data Explorer: Restructured and Redesigned to Better Reflect Data Relationships

Research data is being produced at a rapidly increasing rate, and we at the Department of Energy (DOE) Office of Scientific and Technical Information (OSTI) recognize the importance of making data findable and accessible and are committed to being active in the data community.  The DOE Data Explorer (DDE), launched in 2008, is a search tool enabling users to locate and access research data resulting from DOE funding.  Additionally, OSTI joined DataCite in 2011 and established the DOE Data ID Service, through which OSTI assigns persistent identifiers, known as digital object identifiers (DOIs), to datasets submitted by DOE and its contractor and grantee researchers to help increase access to digital data from DOE-funded research.

As product manager for DDE, my goal is to ensure that DDE reflects the ways in which researchers organize their data so that users can easily find or discover data of interest to them.  So, after OSTI held meetings with researchers and stakeholders and listened to what they need in a data search and discovery tool, we developed a plan to reorganize and redesign DDE to better reflect inherent relationships between data objects.  I presented this plan at the SciDataCon conference in September 2016 and subsequently co-authored a paper with colleagues Carly Robinson and Jannean Elliott that further explained the steps being taken to update DDE.

Through a multi-phase restructuring, we are creating meaningful contextual relationships between data objects to make the data more accessible; phase I was completed in August 2017, and phase II is ongoing as we work with our researchers and stakeholders to make further improvements to reflect how they organize and search for data. 

Previously, DDE had two product types, Datasets (single instances of data whose boundaries have been defined by the data client, with an associated DOI) and Data Collections (records created and curated by OSTI that describe a data project funded by DOE).  However, from our researchers and stakeholders, we learned that a third record type, data Project, was needed to represent a top tier or a level broader than our current definition of a Data Collection. 

In restructuring DDE, we recast the data Collection as a middle tier, with data Project representing a scientific research group, data center, user facility, or other DOE-funded endeavor creating research data.  A data Collection is a package of related datasets with a DOI for the entire collection.  A Dataset is a single instance of data whose boundaries have been defined by the data submitter, with an associated DOI.  The data is not hosted at OSTI but resides at national laboratories, data centers, user facilities, colleges and universities, or other websites.

These three types of records represent the various levels that occur during data creation and use, allowing you to search and discover data through the natural hierarchical relationships that exist between data products.  Now, when you perform a search in DDE, your results will be broken out by Projects, Collections, and Datasets, all navigable by tabs at the top of the search results page.

Selecting a Project record will take you to the full citation page, where you can see information about the overarching project, as well as a list of all related datasets submitted to OSTI from that project.  This helps to increase discoverability by associating the datasets with the overarching project and providing additional context for the creation and use of the data. 

My OSTI colleagues and I are excited about the new search and discovery features offered in DDE.  To create data Collections, we are working with researchers who already have obtained DOIs for their Datasets and specifically requested this hierarchy.

We appreciate input from the data science community in requesting this search structure.  As we continue our work to improve how DDE works for researchers in phase II, we invite your feedback.  For questions or comments, or if you would like to learn more about DOE Data Explorer or the DOE Data ID Service, please contact us at ddecomments@osti.gov.

Sara Studwell is a Librarian and Product Manager for DOE PAGES, DOE Data Explorer, and DOepatents in the DOE Office of Scientific and Technical Information.