alt Hacker NewsMusic recommendation is such a hard problem. There are all these seemingly obvious relationships you can map between bands to create a big graph that looks good but that almost never captures what goes on when a human with deep music knowledge recommends music. Often the best recommendations have no obvious relationships to the band you like.
I played around with this tool a bit and didn't find it any better then other more traditional music discovery tools, that is to say not very effective.
For example, I entered John Zorn and was recommended a bunch of John Zorn's bands. I entered The Residents and got The Pixies :/
I think its more effective to click around on Music Brainz and Wikipedia.
Replies
the problem is there's different ways that people engage with music. Some listen to the lyrics and want to have an emotional connection, some view it as exploratory art, others wear it as an identity, some are just looking for similar sounds ... You need to have a routing system that can match the recommender to the style of engagement.
If you don't have that, you can't solve it.
You seem knowledgeable about this.. Care to test my old project for music recommendation? I built it by looking at co-occurrence of artists in Spotify playlists, which gives me word2vec-style vectors, and then its just kNN.
No login needed, just enter some artist names and see what you get:
Nothing beats humans with great music tastes and deep knowledge. I’ve yet to find any form of recommendation engine that has surprised and delighted me the way humans have.
This tool might unearth something interesting, but I find it sus that it’s recommended the same artist (Adrianne Lenker) when I asked about Aimee Mann and Steven Jessie Bernstein.
Pandora solved this problem nearly 20 years ago an Spotify with all its money and engineers do such a bad job, it’s beyond absurd.
If you're into John Zorn and The Residents, you gotta check out Angine de Poitrine: https://www.youtube.com/watch?v=0Ssi-9wS1so
Microtonal polyrhythmic looping absolute madness. (you can hear some Primus and King Gizzard and the Lizard Wizard kinda sounds in there, if they also tickle your fancy)
Residents -> Pixies is certainly an odd recommendation. Having said that, where _can_ you go from The Residents? Daniel Johnston?
Spotify seems to have mastered music recommendation.
It would be great if somebody could reverse engineer their recommendation algo
For me music discovery is a solved problem. Here's the algorithm:
1) Imagine the timeline of musical history. If you don't have a clear idea of it, Wikipedia is a good place to start.
2) Pick any genre/period you don't know well. (For example, medieval music, or swing-era jazz.)
3) Look up the main figures of that genre/period. (For example, Guillaume de Machaut, or Duke Ellington.) Wikipedia is good for this too.
4) Listen to a sample of their most well known pieces. YouTube is good for this.
5) Repeat. Go down rabbit holes when you like.
No fancy tools needed, just your mind and the internet. This will give you interesting music for many years, and improve your musical taste a lot too.