Posted on 02 December 2012

Part 4 in a series of videos recorded from ACM MIRUM 2012 in Nara, Japan.

Ordinary music recommendation services provide recommendations based upon a user's profile but without any knowledge of the context (e.g. mood, location, activity) in which the user is listening to the music. In this talk, Marius Kaminskas presents a method for music recommendation which incorporates location information to provide recommendations that are more appropriate for the user's location, for example, music that is more culturally or emotionally connected to a famous city or tourist attraction.