Submissions:2018/automated recommendations for sources
This submission is waitlisted for WikiConference North America 2018. It was not scheduled in the first round of scheduling, but may still be scheduled after further review, as slots open up, or as a pre-conference session.
- Title
- Theme (optional)
- Tech & Tools
- Type of submission
- presentation
- Author
- Lane Rasberry
- E-mail address
- rasberry
virginia.edu
- Wikimedia username
- user:bluerasberry
- Affiliation(s) (optional)
- University of Virginia
- Abstract
Can artificial intelligence do automated writing of Wikipedia articles? No, perhaps machine learning can make recommendations to match published reliable sources to Wikipedia articles which could summarize and cite them. For example, if there is an article about a particular medical condition, could there be an automated process for a bot to recommend that article on the talk page of the relevant Wikipedia article?
Cochrane is a nonprofit organization and publisher of academic systematic reviews of medical guidelines. The Wikimedia community has recommended the articles as reliable sources for medical content since at least 2010, yet human labor has not yet matched many thousands of their publications to the Wikipedia articles where a wiki editor could summarize them.
This talk will give a brief overview of the quality control process in Wikipedia, a description of source citations, and present the Cochrane Collaboration Article and citation matching tool development project.
- Length of presentation
- 15
- Special requests
- slides
- Preferred room size
- 25
- Have you presented on this topic previously? If yes, where/when?
- no
- If you will be incorporating a slidedeck during your presentation, do you agree to upload it to Commons before your session, with a CC-BY-SA 4.0 license, including suitable attribution in the slidedeck for any images used?
- yes
- Will you attend WikiConference North America if your submission is not accepted?
- yes
Interested attendees
If you are interested in attending this session, please sign with your username below. This will help reviewers to decide which sessions are of high interest. Sign with four tildes. (~~~~).