Crawl play lists data from xiami. Using these lists data to train recommendation model.
The main script to crawl data from xiami is in python directory. The main python source file is named "xiami.py". You can have a look at it and get to know the structure of data output.
colls.txt {colls_id : [[song_id, artist_id], [song_id, artist_id], ...], ...}
arts.txt {artist_id : artist_name, ...}
songs.txt {song_id : song_name, ...}
colls_name.txt {coll_url : coll_name, ...}