Name		Name	Last commit message	Last commit date
doc		doc
gephi		gephi
graph_graphmls		graph_graphmls
graph_jsons		graph_jsons
graph_pickles		graph_pickles
images		images
paper		paper
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
artistvis.html		artistvis.html
centrality.py		centrality.py
cliques.py		cliques.py
data.txt		data.txt
graphify.py		graphify.py
graphviz.py		graphviz.py
hits.py		hits.py
index.html		index.html
metrics.py		metrics.py
musicdinet.py		musicdinet.py
musicnet.py		musicnet.py
pickle_to_graphml.py		pickle_to_graphml.py
pickle_to_json.py		pickle_to_json.py
pickle_to_json_no_img.py		pickle_to_json_no_img.py
rankArtists.py		rankArtists.py
vincentGraph.py		vincentGraph.py
visualization.html		visualization.html

Repository files navigation

A musical collaboration graph

Understanding musical collaborations via graph-based analysis

Visualization

Root dir: https://akashlevy.github.io/Musical-Collab-Graph/

graph.pickle : All scraped nodes
graph2.pickle : Nodes with degree > 1
graph3.pickle : Only Nodes actually scraped by algorithm
graph4.pickle : Scraped nodes with degree > 1

Dependencies

Methodology

Song data is fetched from Spotify's web API

Graph analysis

(Erica) Calculate pagerank importance vectors, digraph based analysis
(Daniel) Hub/centrality measures, clustering coefficient
(Vincent) Metric based influence: Most collabs,
(Akash) Clique analysis / Almost fully connected

Visualization

Neo4J
(Akash) Options to search up selected cliques
(Sunny) Distance based plot from a starting node

About

Using Spotify's API, we created a massive digraph of musical collaborations between current top Pop artists. Using graph centrality measures and ranking algorithms such as PageRank and HITS, we measured the relative influence of each artist based on their importance in the musical collaboration network.

akashlevy.github.io/Musical-Collab-Graph/index.html

0 stars

1 watching

Report repository

Releases

Packages

Contributors

Languages