This is the repository for analysing the networks of artists who appear in Spotify charts. Please read the following to gain a first overview of the GitHub repo.
Data preparation in R
API access for genre information
API access for popularity information
First analysis (general)
Penalized regression
ANOVA
Min-max transformation
Boxcox transformation
Export graphs and visualisations as imgs
Export graph objects for Gephi
- set up project infrastructure (GitHub repository + local machine)
- Data preparation
- Data collection (Kaggle)
- Data preprocessing
- Transform network object
- Analysis
- Exploratory data analysis
- Hypothesis 1
- Calculate centrality measures
- Perform regression (dependent variable = streams; independent variables = centrality measures)
- Hypothesis 2
- Create genre subgraphs
- Calculate centrality measures
- Perform ANOVA on avg number of streams between different genres + whole network
- Write report