When I started writing Record Store I was interested in providing human curation of music in the face of so much technology in the music space. As a writer and musician, I brought my avid music cataloging to the table, and people responded to the inexplicable human touch of a curated playlist. What most people don't know is that I'm also a Data Scientist with a love for visualization, trend spotting, and tech. After graduating from UC Berkeley in 2023, I couldn't figure out how to combine my interests in human curation and algorithms, which seemed to be totally opposed. Finally, I came up with the idea for turntable, a Record Store recommendation service powered by machine learning. I started Record Store with the intention of introducing people to new music. I can only write so much every week, and the writing is only a conduit for someone trying to hear more of what they like from me. With turntable, you can get recommendations from my personal library.
It's pretty simple right now: run the ml.ipynb notebook. input a playlist of mine you like. You can find those here. The code will spit out a similar playlist for you to try.