View source for Submissions:2019/Building a Scholarly Knowledge Base with Wikidata: A Case Study from Vanderbilt University

{{WCNA 2019 Session Submission
|theme=Relationship Building & Support<br />+ Tech & Tools<br />
|type=Panel
|abstract=Can Wikidata provide an alternative to proprietary research information systems for evaluating scholarly production and impact? The panelists, who are all librarians at Vanderbilt University, explore the potential of using Wikidata to collect and track the disparate research outputs of faculty and staff at a Carnegie Research-Intensive (R1) university.   

The effort to use linked data to measure scholarly productivity is not new. Through the activities of a library-sponsored linked data working group over the past several years, members of the panel have learned to create linked data by generating resource description framework (RDF) serializations from simple tabular files and to use those data for tracking research outputs by querying them with the SPARQL query language. While they developed the capacity to perform those queries programmatically, the overhead of writing data in RDF, loading it into an RDF triple store, and querying it with SPARQL proved too cumbersome for practical daily use.    

Inspired by the use of Wikidata for scholarly communications at Indiana University–Purdue University Indianapolis (IUPUI), the panelists investigated Wikidata as an alternative to generic linked data tools. By turning to Wikidata, the panelists could apply what they had learned from their work with linked data in a more user-friendly environment. By using Wikidata as a hub for scholarly data, librarians at Vanderbilt can now programmatically access and collect information about Vanderbilt researchers from ORCID and about their publications from DOI registry services like CrossRef. The panelists are currently working towards partially or fully automating the aggregation of those data with existing Wikidata items, then transforming those tabular data to be consistent with the Wikibase data model so that new data can be uploaded to Wikidata using tools like OpenRefine and software bots.  

The librarians on this panel will speak from their practical experience using linked data principles for tracking scholarly research and share how the decision to build their workflows around Wikidata has increased their productivity while also making their activities more accessible and comprehensible to colleagues. This panel will be of interest to librarians and higher educational professionals interested in using Wikidata to measure and evaluate scholarly productivity. 

Panelists:

* Clifford B. Anderson, Vanderbilt University (user:Clifford_Anderson) 

* Steve Baskauf, Vanderbilt University (User:Baskaufs) 

* Elisabeth Shook, Vanderbilt University (user: Eshook)
|academic=Yes
|author=Clifford B. Anderson, Steve Baskauf, and Elisabeth Shook
|email=clifford.anderson{{@}}vanderbilt.edu
|username=Clifford Anderson
|affiliates=Vanderbilt University Library
|time=30 minutes
|size=30 people
|requests=None
|presented=No
|present-other=No
}}