Learning about Machine Learning with CRIM

Abstract

In this tutorial-essay I will consider how we can use machine learning, specifically dimensionality reduction and embedding methods, with the CRIM corpus. The guiding question is how style can be modeled quantitatively. Building both on music-theoretical conceptualization and machine learning techniques, it will be demonstrated that unsupervised clustering can serve to some degree as a proxy for stylistic similarity. The CRIM data set provides an ideal case study that will also point to some shortcomings of the computational methodology that can only be resolved by a critical view, drawing on musicological expertise and close-reading of sources.

Date
Oct 20, 2022 — Oct 22, 2022
Location
Haverford College, Department of Music
370 Lancaster Ave, Haverford, PA 19041
Fabian C. Moss
Fabian C. Moss
Research Fellow in Cultural Analytics

Fabian C. Moss is a Research Fellow in Cultural Analytics at University of Amsterdam (UvA). He was born in Cologne, Germany, and studied Mathematics and Educational Studies at University of Cologne, and Music Education (Major Piano) and Musicology at Hochschule für Musik und Tanz, Köln. He obtained is PhD in Digital Humanities from École Polytechnique Fédérale de Lausanne (EPFL). Working with large symbolic datasets of musical scores and harmonic annotations, he is primarily interested in Computational Music Analysis, Music Theory, Music Cognition, and their mutual relationship.