Genre-based Analysis of Social Media Data on Music Listening Behavior

It is frequently presumed that lovers of Classical music are not present in social media. In this paper, we investigate whether this statement can be empirically verified. To this end, we compare two social media platforms — Last.fm and Twitter — and perform a study on musical preference of their respective users. We investigate two research hypothe-ses: (i) Classical music fan are more reluctant to use so-cial media to indicate their listing habits than listeners of other genres and (ii) there are correlations between the use of Last.fm and Twitter to indicate music listening behav-ior. Both hypotheses are verified and substantial differences could be made out for Twitter users. The results of these investigations will help improve music recommendation sys-tems for listeners with non-mainstream music taste