Characterizing user Popularity Preference in a large-scale online video streaming system

Most services present a uniform ranking list according to video popularity in order to satisfy users´ potential demands. But recent work claimed that users in video systems showed distinct preferences on popularity of videos. Understanding these preferences has important implications for designing content distribution mechanism and improving recommendation accuracy. Users´ Popularity Preferences (PPs) in online video streaming services, however, have not been well characterized. In this paper, we characterize users´ PPs and analyze their differences based on a large-scale trace of user watching behaviors in one of the largest online video streaming providers in China. We evaluate central tendency, dispersion tendency and asymmetry of users´ PP distributions. Through hypothesis testing, we prove that users´ PP distributions are not homogeneous. Specifically, users prefer videos with different popularity and the widths of their favorite popularity ranges are different. Within a user´s favorite popularity range, she prefers more popular videos.