Page MenuHomePhabricator

Implement "visibility metric" as percentage of orphan articles in a category.
Open, LowPublic

Description

@Aroraakhil and @MGerlach worked on identifying ways to measure and improve the visibility of Wikipedia articles, by looking at one specific category of invisible articles: orphan articles, or articles that are not linked from anywhere else in a Wiki.

Selection, Extent, and visibility of articles are the 3 aspects that were found to be important when thinking about knowledge gaps metrics and ways to describe inequalities on Wikipedia. In the knowledge gap index, each gap (gender, geography, etc) is now measured via

  • a 'selection' metric, the % of articles about each category in the gap, say man/women, different regions, etc
  • an 'extent' metric, the average quality of articles in each category.

We discussed the feasibility of adding a 'visibility' metric to the knowledge gap index, where we measure a gap by calculating the percentage of orphan articles that belong to each category (28% of orphan articles are about women)? It seems that with the current data and available code, there are viable solutions to do this in a relatively short time frame, hence creating this task.

@fkaelin feel free to add relevant links to repos and data.

Details

Other Assignee
fkaelin

Event Timeline

just added a link in the task description pointing to the previous research that identified selection, extent, and visibility as 3 of the most relevant aspects for metrics for knowledge gaps. https://meta.wikimedia.org/wiki/Research:Developing_Metrics_for_Content_Gaps_(Knowledge_Gaps_Taxonomy)#Outcomes

fkaelin updated Other Assignee, added: fkaelin.

@fkaelin @XiaoXiao-WMF this is a task for research engineering. If we don't have any plans to implement this in the next 6 months, we should probably move this to Research-Freezer ?

yes, please @Miriam unless there is some urgent request from you or Martin.