Slashdot Mirror
Machine Learning and MP3s
dan moore writes "Students at Caltech and Harvard have developed a system that analyzes playlists and learns people's listening patterns. It then channels its knowledge into generating streams of music that the people themselves would like to listen to. Intuitive, accurate, and finally someone has done it. Check out the website to get one of the available plugins. Another interesting approach to digital music."
The problem I see with The Brain is that it doesn't seem to factor out the correlation between music availability and the number of times it's downloaded. Hence, our collective "tastes" are based on what MP3 providers are pushing down our throats.
No, because this isn't based on downloads. Or on collective taste. You play songs in your mp3 player, it pays attention to what you listen to and when, and tries to figure out what you want to hear. Simple idea, shame you missed it.
If you ran The Brain on only MP3.com Top Playlists, I suspect only the sponsored songs would make it to the final stream selections.
Huh? Well, maybe in your world, but back here in the land of people who read shit before they spout off we noticed that, hey, that'd only happen if you were a dumbass and consistenly played the sponsored songs. If you don't play the song, neither does the Brain.