Recommender Systems: Behind the Screen (edX)

You will be accompanied throughout and given concrete examples by seven international experts from both Academia and Industry.
Recommender systems are algorithms that find patterns in user behaviour to improve personalized experiences and understand their environment. They are ubiquitous and are most often used to recommend items to users, for example, books, movies, but also possible friends, food recipes or even relevant documentation in large software projects, or papers of interest to scientists.
The content of this MOOC is an introduction to the field of recommender systems. The outline includes: machine learning for recommender systems followed by an introduction to evaluation methods; advanced modelling; contextual bandits; ranking methods; and fairness and discrimination in recommender systems.
The course is primarily intended for industry professionals and academics with basic (first-year undergraduate) knowledge in mathematics and programming (ideally Python). Graduate students in science and engineering (mainly those who are not yet familiar with machine learning and recommender systems) may find this content instructive and compelling. The content of this course will also be of great use to whomever uses or is interested in AI, in any other way.
We estimate that it takes 6 weeks to follow this class. The course is divided into relevant segments that you may watch at your own pace. There are comprehensive quizzes at the end of each segment to evaluate your understanding of the content. You will also practice recommender systems algorithms thanks to a tutorial guided by an expert. Also, a second self-practice module will be offered to participants who will register for the course with the Verified Certificate.

What you'll learn
At the end of the MOOC, participants should be able to:

• Understand the basics of recommender systems including its terminology;
• Identify the types of problems and the recommender systems’ methods to solve those;
• Apply the methodology for carrying out a project in recommender systems;
• Use recommender systems’ algorithms through practical and tutorial sessions.

Syllabus

Module 1
Machine Learning for Recommender Systems
Score Models
Practical Aspects
MODULE TUTORIAL Matrix Factorization

Module 2
Evaluations for Recommender Systems
Offline (Batch) Evaluation
Online (Production) Evaluation

Module 3
Advanced modelling
Extending Basic Models
A missing Data Perspective
MODULE SELF-PRACTICE Autoencoders (this module is assessed and offered only to participants who register for the course with the Verified Certificate)

Module 4
Contextual Bandits
Introduction to Bandits
Putting it All Together

Module 5
Learning to Rank
Learning to Rank with Neural Networks
Learning to Rank with Deep Neural Networks

Module 6
Fairness and Discrimination in Recommender Systems
Algorithmic Fairness
Fairness in Information Retrieval