Dear all,
In a cooperation project between BFH, SWITCH, and Wikimedia Sweden, we
are currently enhancing the ISA Tool
<https://commons.wikimedia.org/wiki/Commons:ISA_Tool> by providing
tagging recommendations based on a machine vision algorithm and an
algorithm designed to extract entities from metadata on Wikimedia Commons.
For those of you who do not know the ISA Tool yet: It is a crowdsourcing
tool used on Wikimedia Commons to add "depicts" statements to images. It
thus helps the community to add structured data to images, by creating
links between images and related concepts in Wikidata.
The enhanced tool is currently in the beta-testing phase. You are
invited to try it out and to give feedback. In the current test round,
we are focusing on usability and the look and feel of the enhanced tool.
In a second test round, we will explore the notion of what constitutes
"good tagging" and to what extent the algorithm-based recommendations
help users follow good tagging practices.
We deeply appreciate your feedback, since it will help us determine the
next steps of the project.
Please sign-up to indicate interest, test the tool, and simply provide
feedback in the talk page.
More information on the new feature and test campaign:
https://commons.wikimedia.org/wiki/Commons:ISA_Tool/Image_to_Concept
Thank you in advanceǃ
Beat and the team