The only way I've seen of identifying 'gendered' content is to find an
external dataset where you know that people of different genders have
explicit preferences for one item over another, and then to try to map that
onto Wiki[pedia, data] content.
The one example of this approach of which I'm aware is a study done by the
University of Minnesota
<http://files.grouplens.org/papers/wp-gender-wikisym2011.pdf>, where they
compared the size/quality of Wikipedia articles about movies to an external
dataset of movie reviews (provided by users of their own MovieLens
product), where the gender of the people who rated those movies was known.
This approach should be possible, but it's hard to find a good comparison
dataset that's openly available. For example, if you had access to the
(declared) gender of people on Pinterest who pinned certain items, you
could potentially use that dataset for comparison. But I expect that
Pinterest keeps their user data secret--primarily because that's how they
There may be other datasets out there, but I can't think of any offhand :/
On Sat, Feb 6, 2016 at 12:45 PM, Stas Malyshev <smalyshev(a)wikimedia.org>
Is there any way to find out if a clothing item
(dress, hat, shoes, etc)
is gendered? Any ideas how to do this? Thanks in advance
I don't think we have currently anything, we could have something like
"subclass/instance of" "male clothing" or "female clothing"
things are very culture-specific (including changing within the culture
with a passage of time) so I'm not sure it would be easy to represent
this in Wikidata.
Wikidata mailing list
Jonathan T. Morgan
Senior Design Researcher
User:Jmorgan (WMF) <https://meta.wikimedia.org/wiki/User:Jmorgan_(WMF)>