Is there a more efficient way to implement RecSys?

Based on what’s been told in the lecture, we have to train a new model for each user.

To do this more efficiently, we could reduce the number of movies based on the genera of already rated movies and the number of users based on similar interests; but is there a way to train a model based on all the features and users, once and for all, and recommend movies based on the big optimized parameters that we’ve got?