edublancas / song-lyrics

Exploratory Analysis of 200K+ song lyrics from the 1 million songs dataset
https://blancas.io/song-lyrics/
MIT License
5 stars 1 forks source link

Song clustering #4

Closed edublancas closed 6 years ago

edublancas commented 6 years ago

Another way of finding patterns in the songs is to cluster them, then visually explore each cluster. We can first try with simple clustering methods using sklearn, then move to more sophisticated ones - we can use the bag of words representation (cut up to n words) for this. Let see what we get

PCA can help to visualize clustering results