MusicGraph API Documentation

Playlist Overview

The Playlisting API allows you to create powerful DMCA-compliant music recommendations based on hundreds of thousands of potential “seed artists”.

One important differentiator in the way we generate our recommendations is that we use very powerful “graph algorithms” to select the best songs in a real-time for each user, which can take into account a large number of variables, such as musical similarities, personal taste, social relationships, market metrics, user context, etc.

These “graph recommendation” approach not only results in much more personalized selections, but also in a more transparent experience for the end-user, as each recommendation comes with “evidence” that clearly explains why each song is selected. We can achieve this by “walking the graph” and taking note of each step taken in the decision process. What is even cooler is that depending on the user feedback, we can modify the “walk”, so that for example we can skip the “social nodes” in the event that the user dislikes his or her friends’ taste in music.

Right now we only have our “Basic Playlist Generator” enabled, but we will add a ton of new features in the coming weeks that will allow you to fully leverage the full power of MusicGraph’s deep music knowledge.