Visualizing networks of music artists with RAMA

Abstract

In this paper we present RAMA (Relational Artist MAps), a simple yet efficient interface to navigate through networks of music artists. RAMA is built upon a dataset of artist similarity and user-defined tags regarding 583.000 artists gathered from Last.fm. This third-party, publicly available, data about artists similarity and artists tags is used to produce a visualization of artists relations. RAMA provides two simultaneous layers of information: (i) a graph built from artist similarity data, and (ii) overlaid labels containing user-defined tags. Differing from existing artist network visualization tools, the proposed prototype emphasizes commonalities as well as main differences between artist categorizations derived from user-defined tags, hence providing enhanced browsing experiences to users.