Identifying Style-Types in a Sample of Musical Improvisations Using Dimensional Reduction and Cluster Analysis

Creativity research examines both the processes and products of creativity. An important avenue for analyzing creativity is by means of spontaneous improvisation, although there are major challenges to characterizing the products of improvisation because of their variable nature. A useful concept missing from the analysis of improvisation is the idea that the products of a corpus can be organized into a series of “style-types,” where each type differs from others in certain key structural features. Clustering methods provide a reliable quantitative means of examining the organization of style-types within a diverse corpus of improvisations. To look at the utility of such methods, we examined a sample of 72 vocal melodic improvisations produced by novice improvisers. We first classified the melodies acoustically using a multidimensional musical-classification scheme, which coded the melodies for 19 distinct features of musical structure. We next employed multiple correspondence analysis (a dimensional reduction method) and k-means cluster analysis simultaneously, and obtained 3 relatively discrete clusters of improvisations. Stylistic analysis of these clusters revealed that they differed in key musical features related to phrase structure and rhythm. Cluster analyses provide a promising means of describing and analyzing the products of creativity, including variable structures like spontaneous improvisations.