How does music browsing work? On spotify, you can ask for a list of recommendations based either on your recent listening or a single song, which is then generated by a black box. You then have to listen to each of the songs to see what they're like.

What if there was a better way?

Introducing Clusterfy, a way to browse music visually. A user's spotify playlists are scraped, and then the songs on those playlists are clustered based on attributes from the EchoNest API. Then, new recommendations are generated by finding songs which are close to cluster centers. All of this is displayed using the highcharts javascript library, allowing the user to visually find songs which are similar to one they might feel like listening to.

Winner of the 2015 Pennapps 'Best use of Spotify/Nest Echo APIs' award.

Pennapps link

Github link