How Does Spotify Know Me So Well?

I use Spotify quite a bit, mostly to listen to new music, and occasionally to delve into the back catalogue of artists I like. I also listen to their Discover Weekly playlist, which comes up with a list of songs that Spotify thinks I might like, based on what I listen to. Or at least that's what I thought until I read this article, which explains some of the algorithms that Spotify could be using to pick music for me.

Spotify doesn’t actually use a single revolutionary recommendation model. Instead, they mix together some of the best strategies used by other services to create their own uniquely powerful discovery engine.

To create Discover Weekly, there are three main types of recommendation models that Spotify employs:

1. Collaborative Filtering models (i.e. the ones that Last.fm originally used), which analyze both your behavior and others’ behaviors.
2. Natural Language Processing (NLP) models, which analyze text.
3. Audio models, which analyze the raw audio tracks themselves.

If you're interested in big data and how companies track your data, you should read this.