Network landscape of western classical music

Abstract

The expanding availability of large-scale data is leading to increased opportunities for applying advanced data analysis and modeling methodology to a wide range of problems and systems, allowing us to deepen our understandings and make novel discoveries. In this paper we use the tools of network science to the network of musicians from the western classical music tradition constructed from an extensive data archive of CD recordings. We measure the fundamental characteristics of the network such as the small-world property and the power-law degree distribution. We also investigate the community structures of the musicians, revealing how individual attributes such as musical style, positions, and nationalities factor into the large-scale association patterns of the network. Finally, we conduct a more specialized analysis to find closeness or the relevance between musicians - allowing us to understand the egocentric network of the musicians. We believe that our work showcases the potential benefits of network science in the study of arts and humanities.