Academic Year 2022-2023



Kevin Roitero
Unit Credits
Teaching Period
Second Period
Course Type
Prerequisites. Basic knowledge of Programming, Algorithms and data structures, Linear algebra, Probability.
Teaching Methods. Lectures and seminars on specific topics.
Verification of Learning. Oral exam plus optional extra activity on a specific topic (seminar, project, etc.) to be agreed with the lecturer. Possibility of written exam.

Modifications for non-standard situations (e.g., Erasmus students) are possible, but have to be discussed and arranged with the instructor.

More Information. Material (slides, etc.) will be provided using the e-learning platform.
Recommender systems are methods devoted to the creation of personalized recommendations to users. They are used in the context of books, movies, music, e-commerce stores, etc., and they received lot of attention (both from academia and industry) due to the popularity of recommendation and streaming services.

The course aims at presenting the main and most important concepts related to recommendation systems.


– introduction

– history ed evolution of recommendation systems

– neighborhood based collaborative filtering

– model-based collaborative filtering

– content based methods

– knowledge based methods

– hybrid methods

– evaluation

– context based methods

– structural methods

– advanced topics (learning to rank, multi-harmed bandits, graph models, neural models, counterfactual evaluation)

– case studies and specific topics

– “Recommender Systems: The Textbook” –

– “Recommender Systems Handbook” –

– “Practical Recommender Systems” –

– “Recommender Systems: An Introduction” –

– “Dive into Deep Learning” –

– additional material presented by the instructor