Un metodo quantitativo per distinguere stili musicali

In this article a study is described in which the possibilities of Machine Learn-
ing for musical style recognition are explored. These methods can be used to dis-
solve authorship problems or to get more insight in particular musical styles. A da-
taset with compositions of Bach, Haendel, Telemann, Mozart and Haydn is inves-
tigated. A number of features (style markers) is measured in the scores. It is shown
that with these features it is very well possible to recognize the styles of the five
composers automatically using various Machine Learning algorithms.