Music Graph library

• abstracts music library graph for any type of data and sources.
• intended use: recommending system helper, and data visualization of personnal library.

What's done ✔️

• data fetcher for spotify API using spotipy
• data fetcher for tidal API using tidalapi

TODOs

• base the graph implementation from the naive current implementation to boost/networkx graph in order to use existing fast graph algorithms.
• merging different APIs graphs
• clustering of graph to create playlists and visualisation
• recommendation based on extracted features (based solely on graph nodes and relationship + metadata)
• recommendation based on textual features (artist bio + playlist description + lyrics)
• recommendation based on musical features (music sample)

Idea

1st Step: Graph creation Possible 2nd steps:

• clustering to create playlists
• recommendations using that graph