Analysis of a Spotify Collaboration Network for Small-World Properties

TL;DR

This paper examines the small-world properties of a Spotify artist feature collaboration network, focusing on clustering and diameter, and Louvain community detection reveals distinct collaboration clusters aligned with genre-based and industry-driven connections.

Abstract

This paper examines the small-world properties of a Spotify artist feature collaboration network, focusing on clustering and diameter. We analyze the giant component and subgraphs based on genres, country-specific charts, and detected communities to assess their small-world characteristics. Results indicate that the network is scale-free and follows a power-law degree distribution, with highly popular artists serving as central hubs. Louvain community detection reveals distinct collaboration clusters aligned with genre-based and industry-driven connections. These findings offer insights into music recommendation systems and digital collaboration trends, contributing to a broader understanding of artist networks in the digital age.

Figures (34)

• Figure 1: A visualization of the disconnected graph.
• Figure 2: A graph of the top 0.1% of artists based on number of connections.
• Figure 3: Comparison of network metrics for the chart-topper subgraph, a comparable random Erdős-Rényi graph, and a comparable regular lattice.
• Figure 4: Histogram of the degree distribution.
• Figure 5: Log-log plot of the degree distribution with the fitted power law line.