All course materials, announcements, information and links for all students will be provided on Moodle


DayTimeFrequencyPeriodRoomLecturerRemarksMax. participants
Vorlesung(V) - Lecture - Dates/Times/Location:
Mon. 13:00 bis 15:00 weekly 04.04.2022 to
G40B-238 (100 Pl.) Spiliopoulou   100
Übung (Ü) - Exercise - Dates/Times/Location: Group 1
Tue.11:00 bis 13:00weekly05.04.2022 to
G22A-217 (40 Pl.) Tutor


Übung (Ü) - Exercise - Dates/Times/Location: Group 2
Wed.13:00 bis 15:00weekly06.04.2022 to
G22A-203 (40 Pl.) Jamaludeen



Overview (from LSF)

Learning Content

Outline of the course

Part I - Basics of recommenders:

  1. Formalization of the recommendation problem
  2. Rudimentary best-guess solutions
  3. Evaluating a recommender in the lab

Part II: Neighbourhood-based learning methods

  1. Learning on a matrix of 0/1 ratings
  2. Collaborative filtering
  3. Methods based on matrix factorization
  4. Learning to predict (the future)
  5. Outlook for the cold-start problem

Part III: Exploiting text content

  1. Basics on learning from texts
  2. Content-based recommenders
  3. Deriving ratings from opinions &  Opinion mining

Part IV: The surroundings of a recommender

  1. Recommenders in the context of Customer Relationship Management
  2. Trusting a recommender

Part V: Study of scientific papers – compulsory for the 6 ECTS version, optional for the 5 ECTS version


Recommended background
Bachelor students: You should have completed the compulsory CS courses of the early semesters.

All students: Having attended data mining or machine learning or an AI course is of advantage, although you may catch up with home reading. You must refresh your secondary school background in statistics.


Goal of this course is to make you familiar with recommenders. You will learn what requirements are placed to a recommender by the business operating it and by the users interacting with it, and you will become proficient in methods used to meet these requirements.

A recommender has a front-end service responsible for the interaction, and a back-end machine learning core that derives the recommendations to be presented to the users and learns from the users' behavior. Most part of the course is on the back-end: you will become familiar with basic and advanced methods that model the recommendation task as an optimization problem and deliver solutions for it.

A recommender learns from information on the items to be recommended and the users to be served. You will see methods that extract such information from data – mainly opinionated texts.


Book of the course:

  1. Introduction to Recommenders Systems Handbook, SPRINGER

 Core papers on the underpinnings of specific course topics:

  • J.L. Herlocker, J.A. Konstan, L.G. Terveen and J.T. Riedl (2004) Evaluating Collaborative Filtering Recommender Systems, ACM Trans. On Information Systems, Vol. 22, 5-53, 2004 (on: Evaluation)
  • M. Hu and Bing Liu (2004) Mining and summarizing customer reviews, Proc. of ACM Int. Conf on Knowledge Discovery from Data (KDD'04), Test-of-time Award at KDD'15 (on: Opinion Mining)
  • G. Takács , I. Pilászy, B. Németh and D. Tikk (2009) Scalable Collaborative Filtering Approaches for Large Recommender Systems, Journal of Machine Learning Research 10, 623-656 (on: CF & MF)

Further papers are cited in the materials of the course.

The additional materials for Part V will be announced during the course.



The course RECSYS can be examined for 5 ECTS or for 6 ECTS.


To acquire the 5 ECTS, enroll for the 5-ECTS-exam.


For the 6th credit point, you make an additional assignment which involves homework. There are two options for this assignment:

  • Option DEFAULT: You read one-two designated scientific papers discussed in class for Part IV. The titles of these papers will be announced in class, when Part IV starts. In the exam, you must answer arbitrary additional questions on this/these paper(s).
  • Option CHOICE: You read one scientific paper from a choice of papers on Part IV, make an in-class presentation on specific questions for this paper, and compile a short report on  the paper, including the answers you presented and the feedback you received. The titles of the papers offered for this option will be announced and distributed in class, when Part IV starts. During the exam, you are called to answer to a choice of the questions from the report you wrote. You can select Option CHOICE only in person, during the Meeting at which the papers are offered (most likely Meeting 10).

For the 6th credit point, Option DEFAULT is assumed for all students who did not actively select Option CHOICE, including those that started with the CHOICE assignment but did not finish it.  

Examination modalities

The RECSYS exam is oral by default. Depending on the number of students attending the course, it may be turned to a written exam (as usual for many FIN courses).

  • Written exam: RECSYS-5ects and RECSYS-6ects take place at the same time and have the same duration (120 min). If you have opted for CHOICE under RECSYS-6ects, you may still ask for the DEFAULT exam sheet. Make sure that you get the correct exam sheet, because no changes are permitted after you got an exam sheet.
  • Oral exam: make clear whether you take the exam for RECSYS-5ects or for RECSYS-6ects. If you have opted for CHOICE under RECSYS-6ects, you may still decide the option DEFAULT:  you must do so before the examination begins.


Target Group<p>RECSYS is for bachelor students in high semesters and for master students, as follows:</p><ul><li>5 ECTS: Bachelor INF, INGINF, CV and Bachelor WIF</li><li>6 ECTS (additional materials): Master INF, INGINF, WIF, DigiEng, Data Science Master DKE</li></ul><p>The exercise class for the bachelor degrees is on German, for the master degrees on English.</p><p>Under Exam you find additional information on how you acquire the 5 ECTS, resp. 6 ECTS <a name="Uebungen"></a></p><p> </p>
Description Recommenders



Last Modification: 05.04.2022 - Contact Person: