Submissions:2021/WikiLetters Systematic Review

This submission is a draft and has not yet been submitted for review.

To the submission author: When you are ready to submit, edit the page and change status=Draft into status=Pending to submit.

Etherpad

Title:

WikiLetters Systematic Review

Theme:

Tech & Tools

Type of session:

Presentation

Abstract:

A "literature review" is the common research practice of survey existing research publications on a particular topic. When researchers conduct a literature review which is reproducible and quantitative, then it becomes a "systematic review". Here we share the WikiLetters Systematic Review tool, which uses Wikipedia platform content and data science technology to quickly identify and profile a set of papers for systematic review.

To operate the tool, the researcher gives the tool about 5-10 papers on a subject, then suggests up to 1000 additional papers collected through traditional pre-data science systematic review techniques. The tool then clusters the papers such that the researcher can quantitatively separate papers into two groups: those meriting inclusion in the review, and those which are out of scope.

This tool imagines the Wikipedia platform as a starting point for scholarly literature review in any field. It builds upon the WikiCite project, which is the Wikidata community effort to open metadata from Scholarly literature, and combines that dataset with contemporary algorithms for analysis and visualization.

While other tools for aiding systematic review exist, benefits of this one include the following: it is free and open; it advances the activism and consumer rights protection of the existing WikiCite project; developing the Wikipedia platform as a center serious scholarly research may be a useful strategic direction; and aside from scholarly use, everyday Wikipedia editors can benefit from casually generating systematic reviews for any Wikipedia article where academic publications cover the subject.

Academic Peer Review option:

Yes

Author name:

Larissa Driemeier, Joseph Harari, Marcos Lopes, Daniel Mietchen, Lane Rasberry, F.P. Andutta

E-mail address:

rasberryvirginia.edu

Wikimedia username:

bluerasberry

Affiliated organization(s):

Polytechnic Institute, Oceanographic Institute, and Department of Linguistics at the University of Sao Paulo; School of Data Science, University of Virginia

Estimated time:

25 minutes

Special requests:

Have you presented on this topic previously? If yes, where/when?:

no

If your submission is not accepted, would you be open to presenting your topic in another part of the program? (e.g. lightning talk or unconference session)

yes