Create a recommendation engine using machine learning.
We have millions of things for someone to listen to. We want to scale to thousands and then millions of users so we need a database that will scale as we grow and be dynamic as we grow. There will be thousands of new episodes to listen to every day so we need to recommend old and new episodes based on their interested.
Their interested will be determined by what they have listened to in the past and what other listen to who listen to similar things. You will help define the attributes that we will use to create the model.
The data must be updated in real time as their are new episodes.
Please let me know what type of database you would use and how you would structure it.