Skip to content

Guess the gender of a person using pronouns from Last.FM and Wikipedia

License

Notifications You must be signed in to change notification settings

stringfellow/genderify

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Genderify

Purpose

To discover the potential bias toward a particular gender of musical artists amongst a person's listening. Using Spotify as a source of artists, and of a person's music library, playlists etc, the biographies (found on either last.fm or wikipedia) of the artists are inspected to see if it is possible to determine a gender - or genders in the case of a group, which for the purposes of this exercise are limited to:

  • Non-binary
  • Female
  • Male

... though the authors respect and appreciate this is an over simplification.

Caveats

There are so many, I'm sure, but here are a few very obvious ones:

  • This work does not serve as a body of facts nor opinions, it is a simplistic algorithmic determination of gender based on the English language, without the nuances of time, intuition, fact-checking, personal communications or anything else that might make it official in any way. It's a dumb robot.
  • We can't always determine gender from a person by their bio - they may privately identify as one thing while publicly identifying another way, for personal reasons.
  • 'Gender' of a band or group is difficult, so we've opted to just represent the split of that group.
  • Groups or artists with the same name will probably confuse this dumb script. It will only store whichever one it encounters first, or which the Last.FM bio lists first when it has multiple.
  • Because it's quite hard to get a list of "all the artists in the world" (is it?) we are just using Spotify's search API which gives an opaquely sorted list of artists back with each call - so, we probably miss a lot.

Conflict

Having just finished reading Cordelia Fine's "Delusions of Gender"1, I (Steve) feel a bit conflicted about this exercise, since it is adhering to a divide which I do not believe should be reinforced, and it seems that explicitly drawing out the genders of artists is doing exactly that (particularly with respect to those many many identities which we've dumped into a single "non-binary").

On the other hand, with a feminist hat on, it feels more important that while the social divide does exist, we ought to do everything we can to redress the balance. So, we offer this piece of work to attempt to lift women and non-binary people out of the shadow of men.

1: A must read. Delusions of Gender, Cordelia Fine, 2010 (W. W. Norton & Company)

About

Guess the gender of a person using pronouns from Last.FM and Wikipedia

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages