The Impact of Context-Aware Recommender Systems on Music in the Long Tail

Music discovery and consumption have changed dramatically in recent years. Nowadays, most people consume music on their personal computers and mobile devices via Internet. Web sites and online services now typically contain millions of music tracks, which complicate search, retrieval and discovery of music. Music recommender systems can address these issues by recommending relevant and novel music to a user based on personal musical tastes. A typical scenario in music recommendation is that the music consumption is focused on some popular tracks and the most unpopular tracks are rarely accessed (i.e., the Long Tail problem). In this paper, we evaluate the impact of using context-aware recommender systems to recommend music in the Long Tail. We perform a set of experiments by applying three different context-aware recommender systems on three different real world data sets. Our findings show that by using contextual information we can improve music recommendation in the Long Tail scenario.