npasslick / spotify-shuffle-analysis

Analysis of the Spotify shuffle algorithm for true randomness

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Analysis of the Spotify Shuffle Algorithm

This repository holds functions and data to analyze the randomness of the Spotify shuffle algorithm. The algorithm is often suspected to be non-random, to favor certain songs in a playlist and to repeat many songs. This statistical analysis should help to better understand the algorithm and prove or disprove the assumptions. This Reddit post might be interesting to read in this context.


Experiments

The analysis is currently a work-in-progress.

Experiment 1

Experiment 1 represents the shuffle behaviour over one complete music library run. If the Spotify music library holds n tracks, the shuffle behaviour of n track skips without interruption is analyzed.

Experiment 2

Experiment 2 represents the shuffle behaviour over multiple, smaller music library runs. If the Spotify library holds n tracks and one run skips m tracks, the shuffle behaviour of n track skips in r = n / m runs with a restart of the playback in every run r is analyzed.


Results of the Analysis

🚀 Coming soon!

About

Analysis of the Spotify shuffle algorithm for true randomness

License:MIT License


Languages

Language:Python 100.0%