Talent scouting in P2P networks

Record labels would like to identify potential artists as early as possible in their career, before other companies approach the artists with competing contracts. However, there is a huge number of new artists, and the process of identifying the ones with high success potential is labor intensive. This paper demonstrates how data mining in P2P networks can be used together with social marketing theories in order to mechanize most of this detection process. Using a unique intercepting system over the Gnutella network we captured an unprecedented amount of geographically identified queries, allowing us to investigate the diffusion of music related content in time and space. Our solution is based on the observation that successful artists, start by growing a discernible stronghold of fans in their hometown area, where they are able to perform and market their music. Only then they manage to breakthrough to national fame. In a file sharing network, their initial local success is reflected as a delta function spatial distribution of content queries. Using this observation, we devised a detection algorithm for emerging artists that suggests a short list of artists with breakthrough potential, from which we showed that about 30% translate the potential to national success.