Master Seminar "Machine Learning Solutions for Business Practice" - M.Sc.

General Information

  • Module: Machine Learning Solutions for Business Practice
  • UniVZ: TBD
  • Cycle: single-time
  • Credits: 6 ECTS
  • Required prior knowledge: None
  • Recommended prior knowledge: Programming skills, ability to apply machine learning models (preferably taken part in at least one machine learning course)
  • Courses of study: Master WI, Master Ufü
  • Type of teaching and learning: Seminar (2 SWS), groups of 3-4 students, maximum of 14 participants
  • Each seminar paper will be prepared by a group of two to three students. First pre-registered, first served.
  • Examination: Attendance at seminar dates, handing in a seminar paper (max. 8000 words), and holding a presentation (approx. 30 minutes).
  • Language: English
  • Lecturer: Dr. Andre Hanelt
  • Contact: Schahin Tofangchi, Patryk Zapadka

Summer Term 2018 Dates

  • Pre-Registration period: The pre-registration takes place from 16th April to 25th April. Each prospective participant shall write an informal e-mail to Schahin Tofangchi, stating their experience in machine learning, in order to register. E-mails sent outside of this time period will be strictly ignored. Students who have registered early will be prioritised during the topic selection procedure.
  • Registration at the Examination Office: The registration in FlexNow will be open in May 2018.
  • Kick-off meeting: 28 May 2018
  • Handing in the presentation: 14 August 2018
  • Date of presentation: 15 August 2018
  • Presentation time: Each group has to present for 20 minutes, followed by a 10-minutes discussion.
  • Handing in the paper: 31 August 2018


  • Topic 1: Smart Customer Segmentation: Predicting Online Trading Behaviour
  • Topic 2: Predicting Customer Inactivity and Evaluating Counter-Measures
  • Topic 3: [self-imposed task]

Learning objectives

The students will be able to apply a wide variety of algorithms in the field of machine learning and can assess their suitability for different problems, handle large amounts of data and prepare appropriate concepts for their processing. Furthermore, they will acquire in-depth knowledge of the state of the art regarding research on different problems in the field of machine learning. Based on data from the domain of financial services provided by the FinTech Group AG, students will be able to elaborate a research problem in a specific business context and develop an application to solve this problem on the basis of machine learning.

Pay attention to the guidelines for seminar theses.