The plan:
- Use AlbumOfTheYear to determine the top artists of each year, their rank in the list, and the length of the list. We can then determine their relative quality within the list.
- Use the last.fm API (or scrape last.fm if necessary) to determine the members of each band. This part is likely the most difficult and could likely require the use of some NLP.
- Use gendelrize.io (or perhaps some R package) to predict the gender of each name
- Find open-source methods for predicting race based on name.
- Cool visualizations and stuff