Algorithms are everywhere these days. In some ways, they’ve always been in our lives, but the internet age has brought them to the forefront in a way we were never aware of previously.
We all know the rough gist of how they work. If they see you like something, they give you more of it. If you like making music, you’re probably going to get lots of adverts for tracks, sample packs, hardware and so on.
So it was with interest that I read about this study of last.fm listeners. The study concluded the algorithms worked better when you were trying to find mainstream music as opposed to more underground material.
On the one hand, this finding is hardly surprising. Algorithms work best when they have lots of information to work with. Information shows you patterns and indicates behaviour. Underground music, by its definition, is not as listened to so the algorithm has less data to work with.
I would, however, be interested to see this experiment replicated across other platforms. I’d be especially curious to see the results on YouTube specifically, because I’ve always discovered loads of old goodies from the suggestions on there.
Now Spotify? They can do one…