Classification and Comparison of Different Folk Music Traditions Using Self Learning Algorithms
Zoltán Juhász · 2010
The results of a comparative analysis of 25 folk so ng databases are summarized in this paper. The meth od of the analysis was based on self organizing mapping and multidimensional scaling algorithms. The results sho w a clear system of deterministic contacts of the cultures. D e concrete common musical forms of different cultur es can also be studied. Methods The analysis was based on automatic comparison of the melody contour, characterized by equidistant pitch samples (see Figure 1.). The number of the pi tch samples was a pre-defined constant value, independently of the time duration of the individua l songs, and all melodies were transposed to the sa me final tone G [1]. Thus, each melody of our folksong database was described by a multidimensional vecto r, pointing to a point of a multidimensional “melody s pace”. The basic idea of the analysis is that the melodies of a given musical culture construct a spe cial point system in the melody space, and the musi cal principles determining the musical forms in the giv en culture may be mapped into the spatial structure of this point system. In other words, the dense as wel l as sparse areas of the point system can be traced back to musical reasons.