Music information visualization and classical composers discovery: an application of network graphs, multidimensional scaling, and support vector machines

Abstract This article illustrates different information visualization techniques applied to a database of classical composers and visualizes both the macrocosm Common Practice Period microcosms twentieth century music. It uses data on personal (composer-to-composer) musical influences generate analyze network graphs. Data style ‘ecological’ are then combined composer-to-composer build similarity/distance matrix, multidimensional scaling analysis is used locate relative position map while preserving pairwise distances. Finally, support-vector machines algorithm classification maps. falls into realm an experiment in music education, not musicology. The ultimate objective explore parts heritage stimulate interest discovering composers. In age offering either inculcation through lists prescribed compositions explore, or recommendation algorithms that automatically propose works listen next, alternative path might promote active rather than passive discovery their less restrictive way prescription.