i dont know how and what data the algorithm takes into account i dont know if it can be reworked or simple it is what it is
(I’ll preface this by saying I’m not familiar with the algorithm)
a couple possibilities might be the fact that there’s not a lot of listens to the artists you’re expecting to see here, and there’s a lot of data for popular tracks (a lot of which are in English). looking at your Similar users, you’ve got very low percentages (all but one less than 3%), so that might play into it.
even looking at my Created for you playlists, there’s quite a few massive artists in there, even though I don’t feel like I listen to the big artists that much…
1 Like
so you say I am special 
if it looks at similar users for clue it means there is needed for more diversity of listeners or i need to make duplicate accounts
More diverse, but also just more users in general. ListenBrainz just passed 25k users 
But there is still a long road ahead of us. In other words: tell your friends!
3 Likes
Duplicate accounts wouldn’t help as it would be the same music 
You want recommends, it needs more data not just more of the exact same data. I’d suggest getting friends with similar taste to join and submit listens. Or find online communities(Facebook, reddit, etc) for the genres you like and spread the word of ListenBrainz. If you know an influencer, influence them to influence their followers to join! 
Also it goes without saying, but make sure all your listens are linked so the site knows what they are and knows what to recommend.
There will be a new Year In Music for all users after 2023 comes to an end. Share yours on your socials and let everyone else know where to get their own!
3 Likes
I have had issues with recordings with Japanese script being matched correctly so there may be something to this. I haven’t dug into it though.
Made a post on Lemmy: https://feddit.nl/post/7753628
(though I doubt Fediverse folks would diversify LB 
)
7 Likes