LocalifyMusicEvents-USA-2019
is the project repo for the paper Towards Quantifying the Strength of Music Scenes using Live Event Data. The paper has been accepted by the International Society for Music Information Retrieval (ISMIR) Conference 2022.
@INPROCEEDINGS{zhou2022musicevent, AUTHOR = {Zhou, Michael and McGraw, Andrew and Turnbull, Douglas}, TITLE = {Towards Quantifying the Strength of Music Scenes using Live Event Data}, BOOKTITLE = {Proc. of the 23rd International Society for Music Information Retrieval Conference}, YEAR = {2022}, ADDRESS = {Bengaluru, India} }
The LocalifyMusicEvents-USA-2019
dataset used in our paper contains information for 308,051 music events that took place in 2019 and from 1,139 US cities, all of which have a population of 10K or more. This dataset consists of data for 28 socioeceonomic indicators, ranging from 6 different categories: Transportation, Population, Economics, Age, Education, and Race. We also introduce the Live Event Music Rate (LMER), a one-dimensional heuristic for measuring the liveness of a music scene, in this dataset. LMER is calculated as the total number of events during 2019 listed in the city divided by the city population according to 2019 Census population estimates. To use this dataset, download it here.
We explore the statistical correlations between LMER and the 28 socioeconomic indicators in this Jupyter notebook.
We grouped the 1,139 cities by population:
- Small (10K-100K)
- Medium (100K-500K)
- Large (500K+)
The correlation plots generated from our results for each of these categories above, as well as for all cities, can be found in this folder.
To see how our dataset was constructed, check out this Jupyter notebook. IMPORTANT: This notebook scrapes the LATEST data from CensusReporter, as CensusReporter has changed from ACS-2019 to ACS-2020 since our original submission. Do not run this notebook if you intend to reproduce our data and results.
We also have a full dataset consisting of 1,246 cities with all population sizes, which can be found here.
-
Clone this repository:
git clone https://github.com/JimiLab/LocalifyMusicEventData.git
. -
Download the SQL database file from here. Put this SQL file in the folder
LocalifyMusicEvents-USA-2019/data
. -
To reproduce our results used in the paper, run the notebook
LocalifyMusicEvents-USA-2019/LocalifyMusicEvents-USA-2019_Step1_StatistcalAnalysis.ipynb
.
Poster presented in ISMIR 2022 (Bengaluru, India):
Doug Turnbull, Tim Clerico, John Hunter, and Emmett Barry all contributed to the Localify.org music event scraper code. This research was supported by NSF grant IIS-1901330/1901168 and NEH grant HAA-280975-21. Finally, we would like to thank the anonymous ISMIR reviewers for their highly constructive feedback.
This library is released under the terms of the MIT license. Full details in LICENSE
file.