Research:Screening WikiProject Medicine articles for quality

This page documents a completed research project.

Recent work in article quality assessment detection[1][2] can enable us to automatically identify which articles are most due to be re-assessed. Let's apply this method to WikiProject Medicine's stubs.

Methods

edit

Articles by category

edit

A relevant set of stub-class articles can be gathered by scanning en:Category:Stub-Class_medicine_articles.

Determining which articles need re-assessment

edit

Use http://pythonhosted.org/wikiclass. Look for articles that are probably not Stubs.

Results

edit

Out of 8,818 WikiProject medicine stubs in the category, 50 were redirects and removed for assessment (these redirects are listed below). The remaining articles were run through the wikiclass quality predictor.

Predicted class counts

edit

The following table shows the count of the number of articles by predicted assessment class. Note that the "A" assessment class is missing from the quality predictor since the model could not be trained on so few observations.

Predicted class Number of articles
Stub 8278
Start 476
C 14

Sorted prediction table

edit

The following table contains a list of articles with a predicted class greater than Stub ordered by the predicted probability that the real assessment class is greater than stub.

Template:Hidden

Redirects

edit

The following section contains a list of talk pages that are in the Stub-class category, but where the parent article is a redirect.

Redirect list

The articles associated with all these talk pages are redirects, except where noted.

References

edit
  1. Warncke-Wang, M., Cosley, D., & Riedl, J. (2013, August). Tell me more: an actionable quality model for Wikipedia. In Proceedings of the 9th International Symposium on Open Collaboration (p. 8). ACM.
  2. Halfaker, A. & Warncke-Wang, M., Wiki-Class: Wikipedia article quality classification. docs - repo