Modeling Musical Influence with Topic Models

Uri Shalit, Daphna Weinshall, Gal Chechik
; Proceedings of the 30th International Conference on Machine Learning, PMLR 28(2):244-252, 2013.

Abstract

The role of musical influence has long been debated by scholars and critics in the humanities, but never in a data-driven way. In this work we approach the question of influence by applying topic-modeling tools (Blei & Lafferty, 2006; Gerrish & Blei, 2010) to a dataset of 24941 songs by 9222 artists, from the years 1922 to 2010. We find the models to be significantly correlated with a human-curated influence measure, and to clearly outperform a baseline method. Further using the learned model to study properties of influence, we find that musical influence and musical innovation are not monotonically correlated. However, we do find that the most influential songs were more innovative during two time periods: the early 1970’s and the mid 1990’s.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-shalit13, title = {Modeling Musical Influence with Topic Models}, author = {Uri Shalit and Daphna Weinshall and Gal Chechik}, pages = {244--252}, year = {2013}, editor = {Sanjoy Dasgupta and David McAllester}, volume = {28}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/shalit13.pdf}, url = {http://proceedings.mlr.press/v28/shalit13.html}, abstract = {The role of musical influence has long been debated by scholars and critics in the humanities, but never in a data-driven way. In this work we approach the question of influence by applying topic-modeling tools (Blei & Lafferty, 2006; Gerrish & Blei, 2010) to a dataset of 24941 songs by 9222 artists, from the years 1922 to 2010. We find the models to be significantly correlated with a human-curated influence measure, and to clearly outperform a baseline method. Further using the learned model to study properties of influence, we find that musical influence and musical innovation are not monotonically correlated. However, we do find that the most influential songs were more innovative during two time periods: the early 1970’s and the mid 1990’s.} }
Endnote
%0 Conference Paper %T Modeling Musical Influence with Topic Models %A Uri Shalit %A Daphna Weinshall %A Gal Chechik %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-shalit13 %I PMLR %J Proceedings of Machine Learning Research %P 244--252 %U http://proceedings.mlr.press %V 28 %N 2 %W PMLR %X The role of musical influence has long been debated by scholars and critics in the humanities, but never in a data-driven way. In this work we approach the question of influence by applying topic-modeling tools (Blei & Lafferty, 2006; Gerrish & Blei, 2010) to a dataset of 24941 songs by 9222 artists, from the years 1922 to 2010. We find the models to be significantly correlated with a human-curated influence measure, and to clearly outperform a baseline method. Further using the learned model to study properties of influence, we find that musical influence and musical innovation are not monotonically correlated. However, we do find that the most influential songs were more innovative during two time periods: the early 1970’s and the mid 1990’s.
RIS
TY - CPAPER TI - Modeling Musical Influence with Topic Models AU - Uri Shalit AU - Daphna Weinshall AU - Gal Chechik BT - Proceedings of the 30th International Conference on Machine Learning PY - 2013/02/13 DA - 2013/02/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-shalit13 PB - PMLR SP - 244 DP - PMLR EP - 252 L1 - http://proceedings.mlr.press/v28/shalit13.pdf UR - http://proceedings.mlr.press/v28/shalit13.html AB - The role of musical influence has long been debated by scholars and critics in the humanities, but never in a data-driven way. In this work we approach the question of influence by applying topic-modeling tools (Blei & Lafferty, 2006; Gerrish & Blei, 2010) to a dataset of 24941 songs by 9222 artists, from the years 1922 to 2010. We find the models to be significantly correlated with a human-curated influence measure, and to clearly outperform a baseline method. Further using the learned model to study properties of influence, we find that musical influence and musical innovation are not monotonically correlated. However, we do find that the most influential songs were more innovative during two time periods: the early 1970’s and the mid 1990’s. ER -
APA
Shalit, U., Weinshall, D. & Chechik, G.. (2013). Modeling Musical Influence with Topic Models. Proceedings of the 30th International Conference on Machine Learning, in PMLR 28(2):244-252

Related Material