Uncovering success patterns in electronic music

Music has been one of the strongest forms of cultural expression and identity for thousands of years and has evolved into a strongly collaborative artistic domain covering a number of various genres. Here we study the particular genre of electronic music and the world of DJs and producers by analyzing a large-scale musical dataset we collected from free online sources. First, we analyze the dynamics of the annual ranking of the Top 100, and by various statistical tools we capture a threshold rank dividing Djs into long-standing stars and ephemeral artists. Next, we construct the weighted undirected co-release network of these artists, apply backbone filterings on it, and extract its community structure. Here we report that each community typically has one or two leading figures, who used to be No. 1. in the world. By looking at the temporal dynamics of this structure, we observe that the different communities rise, peak, and fall separately over time, that their popularity is driven by their leading figures, and that these figures are amongst the first artists to join their community. We show that a major building force behind such communities is mentorship: around half of the musicians entering the Top 100 have been mentored by current leading figures before they entered the Top 100. We also find that the mentees are unlikely to break into the top 20, yet have much higher expected best ranks than those who were not mentored. This implies that mentorship helps rising talents, but becoming an all-time star requires something more. Our results provide insights into the intertwined roles of success and collaboration in electronic music, highlighting the mechanisms shaping the formation and landscape of artistic elites in electronic music.

Milan Janosov
PhD Candidate, Central European University

My background is Physics, and I am currently a PhD candidate in Network Science at the Department of Network and Data Science at the Central European University. My research covers various topics of network, data, and computational social science, such as network analysis, quantifying success, geo-spatial data, and social media.