This project was made as a final project for Stanford's Machine Learning course, CS229. The program generates a playlist from a seed set of songs selects songs from a given dataset that are similar to the given seed, similar to Spotify's recommended music feature but with a different algorithm (Spotify's algorithm, collaborative filtering, requires a large amount of user data to implement and is therefore impractical). Our algorithm is based on a combination of acoustic spectral data and song metadata, and is explained in more detail along with motivation and results in report.pdf.

Unfortunately, the dataset was too large to upload to Github, so we were unable to create a demo.