**This is an old revision of the document!** ----
====== **Latent Artists** ====== by Ben Swanson and Elif Yamangil Dataset - Assume a fixed vocabulary $V$, which in our experiments is a company internal list of music related multiword terms. Each item $x_i$, from an OO-programming point of view, has the following fields * Artist Name * Echonest ID * Echonest Genres * Treating a sample of reviews for this artist as a bag of words $w \in V$