In this recipe, we will be demonstrating a system that utilizes a known as collaborative filtering. At the core, collaborative filtering analyzes the relationship users themselves and the dependencies between the inventory (for example, movies, books, news articles, or songs) to identify user-to-item relationships based on a set of secondary factors called latent factors (for example, female/male, happy/sad, active/passive). The key here is that you do not need to know the latent factors in advance.
The recommendation will be produced via the ALS algorithm which is a collaborative filtering technique. At a high level, collaborative filtering entails making predictions of what a user may be interested in based on collecting previously known preferences, combined with the preferences of many other users. We will be using the ratings data from the MovieLens dataset and will convert it into input features...