This website requires JavaScript.

Universality of preference behaviors in online music-listener bipartite networks: A Big Data analysis

Xiao-Pu HanFen LinJonathan J.H. ZhuTarik Hadzibeganovic
Dec 2022
We investigate the formation of musical preferences of millions of users ofthe NetEase Cloud Music (NCM), one of the largest online music platforms inChina. We combine the methods from complex networks theory and informationsciences within the context of Big Data analysis to unveil statistical patternsand community structures underlying the formation and evolution of musicalpreference behaviors. Our analyses address the decay patterns of musicinfluence, users' sensitivity to music, age and gender differences, and theirrelationship to regional economic indicators. Employing community detection inuser-music bipartite networks, we identified eight major cultural communitiesin the population of NCM users. Female users exhibited higher within-groupvariability in preference behavior than males, with a major transitionoccurring around the age of 25. Moreveor, the musical tastes and the preferencediversity measures of women were also more strongly associated with economicfactors. However, in spite of the highly variable popularity of music tracksand the identified cultural and demographic differences, we observed that theevolution of musical preferences over time followed a power-law-like decayingfunction, and that NCM listeners showed the highest sensitivity to musicreleased in their adolescence, peaking at the age of 13. Our findings suggestthe existence of universal properties in the formation of musical tastes butalso their culture-specific relationship to demographic factors, withwide-ranging implications for community detection and recommendation systemdesign in online music platforms.