Lecture Notes on Recommender Systems

Modern digital systems provide users with access to an enormous amount of information, products, services, and content. An online store may contain millions of products, a streaming platform may offer thousands of movies and series, and a music service may provide access to millions of multimedia resources. Although having access to such a large quantity of information may seem beneficial, it creates an important problem: information overload.

A user cannot examine every available item. Even when a user knows exactly what they are looking for, searching through a large catalog can require significant time and effort. When the user does not know exactly what they want, the problem becomes even more difficult. This is where recommender systems become useful. A recommender system attempts to answer a question such as: Which items are most likely to be relevant to this particular user?