The Wikimedia Research team
<https://www.mediawiki.org/wiki/Wikimedia_Research> is developing a new
method for automatically recommending sections to add to stub articles.
This method uses machine learning to suggest sections that could be added
to an article, based on the sections that exist in other articles on
similar topics.
We think this method could be useful for helping new editors find useful
onboarding tasks to do. But before we build anything, we need to test the
quality of the recommendations.
We are looking for experienced editors to evaluate these recommendations
and provide feedback to help us improve them. We have built a testing tool
<https://secrec.wmflabs.org> that makes it easy to provide quick,
survey-style feedback on the quality of the recommendations, and we are
also interested in more detailed feedback on the project feedback talkpage
<https://meta.wikimedia.org/wiki/Research_talk:Expanding_Wikipedia_articles_across_languages/Inter_language_approach/Feedback#Feedback_on_specific_articles>.
If you are interested in giving us feedback, please get started by reading
the instructions
<https://meta.wikimedia.org/wiki/Research:Expanding_Wikipedia_articles_across_languages/Inter_language_approach/Feedback#Testing_instructions>
here and then start rating articles!
If you have questions about the project or more general feedback, you can
reach us here
<https://meta.wikimedia.org/wiki/Research_talk:Expanding_Wikipedia_articles_across_languages/Inter_language_approach/Feedback#General_feedback_or_questions>.
Sincerely,
Jonathan Morgan <https://en.wikipedia.org/wiki/User:Jmorgan_(WMF)> & Diego
Saez-Trumper <https://en.wikipedia.org/wiki/User:Diego_(WMF)>, Wikimedia
Foundation Research