Music listening data can be very interesting to analyse when combined with a GPS position (where did the user listen to what song?) and music data (danceability, bpm, mood) from Echonest. It could potentially show my “mood” in certain places and at certain times. With approximate speed from the GPS track you could see if on average I walk faster when listening to high-BPM music.
I’ve aquired my location history (>110000km) of the past 2 years from Google Latitude and exported it with Google Takeout to a JSON file. Getting the Last.fm data from my profile wasn’t as straight-forward, but it worked.
You need to get this Python script, that has been created by Libre.fm so their users can migrate to them from Last.fm:
Make sure you have Python installed and then execute the script like so:
./lastexport.py -u YourLastFmUsername ./scrobbles.txt
No API key or other credential is needed since everyones scrobbles are public. The script downloads all scrobbles with a unix timestamp, song title, artist, album and id into a tab seperated file.