Recommender systems aim to capture the interests of users in order to provide them with tailored recommendations for items or services they might like. Economically, recommender systems are directly linked to revenue and business strategies. In this course, we study traditional as well as state-of the-art recommender systems and prototypical applications scenarios. We also discuss characteristics, limitations, scalability issues and extensions such as personalization.
The students work independently in small groups on a problem that is to be solved with the help of machine learning. A seminar phase takes place first to give the students an overview of current methods of machine learning and data mining. Then, in groups, aspects of the problem will be worked on practically, using current techniques and tools, in order to create and evaluate a common system at the end. The results of the groups will be presented in a short final presentation and individual elaborations.