Jump to main content
Professorship of Media Informatics
Teaching

Teaching

Topics for bachelor's, master's and diploma thesis

On this page you will find various topics for student projects. In the respective fields there is a broad spectrum of topics that can be covered in the context of a bachelor's, master's or diploma thesis and other academic activities, such as seminar papers or internships. In addition, we constantly have numerous other topics at all skill levels which we would be happy to discuss with you.

The following range of topics is available:

There are plenty of guides on how a web application should be designed to have good usability. However, these guides are rarely based on scientific research to support the design recommendations. For this topic, you should take a look at relevant usability websites for recommendations that you should then investigate in your own study, e.g.:

You should think about the following questions before our first conversation:

  • Do you already have relevant research that could be used as a template or replicated? (In physics, a new particle must also be detected at several laboratories before it is considered discovered).
  • What methodological approach is targeted and feasible for the chosen issue?

Relevant publications:

Supervisor: Thomas Wilhelm-Stein (0371/531-37414, E-Mail)

The Data Science portal Kaggle.com offers a variety of competitions and datasets, one of which can be selected and worked on in the thesis.

Please follow the following approach:

  • Select a competition or dataset to work on.
  • Investigate the background and motivation for that competition or dataset.
  • Download the data and explore it.
  • Familiarize yourself with the metrics for evaluation on this data
  • Research possible solutions from the currently available literature
  • Conceptualize and implement your own solution approach
  • Evaluate your approach based on appropriate metrics
  • Compare your approach to previously submitted solutions (e.g., in leaderboards)

The focus is on an analytical approach, not on the ultimate performance or position in the leaderboard.

Supervisor: Stefan Kahl (E-Mail)

This topic deals with the interpretability of artificial neural networks, which are often referred to as "black boxes". In the thesis, different methods from the field of "Explainable AI" will be researched and comprehended in an implementation part. The evaluation is done by comparing different approaches on a sample data set.

Prerequisites for working on the subject are knowledge in the area of Deep Learning and programming with Python.

Supervisor: Stefan Kahl (E-Mail)

Microcontrollers are energy-efficient, but provide only very limited computational resources. This poses a challenge for the use of machine learning methods on such devices. In this topic, methods for increasing the efficiency of deep neural networks will be investigated so that they can be used on microcontrollers.

Prerequisites for working on the subject are knowledge in the area of deep learning and in the development with Arduino or Espressif.

Supervisor: Stefan Kahl (E-Mail)

Culture exerts its influence on many areas. For example, different cultures have different traditions, but also typical clothing and buildings. Is there a comparable connection also with user interfaces?

For inspiration, here are a few examples of contexts that could be explored:

  • Color usage on regional websites
  • Alignment of interactive elements depending on the reading direction (left to right or vice versa)

You should think about the following questions before our first conversation:

  • What connection or aspect of culture and interface design do you want to investigate?
  • Do you already have relevant research that could serve as a template or be replicated? (In physics, a new particle must also be proven at several laboratories before it is considered discovered).
  • Which methodological approach is target-oriented and feasible for the chosen issue?

Relevant publications:

  • Kim, J. H., & Lee, K. P. (2005). Cultural difference and mobile phone interface design: Icon recognition according to level of abstraction. In Proceedings of the 7th international conference on Human computer interaction with mobile devices & services (pp. 307-310). ACM. doi:10.1145/1085777.1085841
  • Jagne, J., Smith-Atakan, A.S.G. Cross-cultural interface design strategy. Univ Access Inf Soc 5, 299–305 (2006). doi:10.1007/s10209-006-0048-6
  • Li, H., Sun, X., & Zhang, K. (2007). Culture-Centered Design: Cultural Factors in Interface Usability and Usability Tests. Eighth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD 2007), 3, 1084–1088. doi:10.1109/SNPD.2007.489

 

Supervisor: Thomas Wilhelm-Stein (0371/531-37414, E-Mail)

 

Each topic has to be presented with a short talk in our professorship meeting prior to the official registration at the examination office. For preparation please use (or follow) this presentation template (TUC Cloud) or this presentation template (Google Drive).