Artist Collaboration Network

Investigating the diversification of musical artist collaborations across genres using Spotify API data

About

Our goal was to answer the question: "How have artists collaborations diversified over time?" With the rise of streaming service through the 2010s, we conduct an qualitative and quantitative analysis to see the trends in a network of artist collaborations over time. This repo is a running record of the data analyses we did. We did so by choosing the top 10 artists from a given genre (Chicago rap), then connecting artist nodes to other artists for a given year. Below is more information about navigating the repository.

Navigation

Generally, the report of all the information can be seen in the report.pdf file. In the main folder is also the five notebooks used for collecting, processing, and analyzing the data. As for the other folders...

• artists and tracks store the raw data pulled from Spotify
• data stores the organized dataframes consisting of tracks info and artist info separately
• gephis and gexf_files are the related network graphs used for analysis and visualization
• figures has all of the plots of quantitative network measure, including degree distribution, assortativity, and average clustering coefficient