Music Mapping: Visualizing the Relationships Between Songs

This project developed a novel method for visualizing the relationships between songs based on how people are arranging them on playlists of audio files. The method is called "Music Mapping", and could serve as a method for exploring new music based on music that an individual already owns. We put together a small movie that explains the results of one such map (the movie file is large... use "right-click, download as" to view). This project proved to be very successful, and the results were presented at a Stanford data mining course as well as at Music 2.0 Conference. This is an ongoing project within a collaboration between MyStrands and the Indiana University.

Justin Donaldson wins the 1st place award in the Netsci06 poster session with ZMDS: Visualizing Structural Entropy in Queried Network Sub-Graphs
[click to enlarge!]
Congratulations Justin!

Case-based Reasoning to Musical Playlist Recommendation

This project implements a Case-based Reasoning (CBR) approach to musical playlist recommendation. Our CBR approach focuses on recommending new and meaningful playlists. The Case Base is formed by a large collection of playlists. Then, the CBR system first retrieves from the Case Base the most relevant playlists, then combines them to generate a new playlist, both relevant to the input song and meaningful ordered. This research is developed in a joint project with the Artificial Intelligence Research Institute of the CSIC.