Seminar Data-Science I (Methods)

CS 721 Master Seminar (M. Sc. Wirt. Inf., M.Sc. MMDS, Lehramt für Gymnasien)

LecturerProf. Dr. Markus Strohmaier, Marlene Lutz
Course FormatSeminar
Credit Points4 ECTS
GradingWritten report with oral presentations 
Examination dateSee schedule below
Information for Students

The course is limited to 15 participants. Please register via Portal2.

Marlene Lutz, M.Sc.

Marlene Lutz, M.Sc.

For further information please contact Marlene Lutz.

Course Information

  • Course Description

    In this seminar, students perform scientific research, either in the form of a literature review or by conducting a small experiment, or a mixture of both, and prepare a written report about the results. Topics of interest focus around a variety of problems and tasks from the fields of Data-Science, Network Science and Text Mining. 

    Previous participation in the courses “Network Science” and “Text Analytics” are recommended.

  • Objectives

    Expertise: Students will acquire a deep understanding of the research topic. He/she is expected to describe in-depth and summarize the topic in detail in his/her own words, as well as to judge the contribution of the research papers to ongoing research.

    Methodological competence: Students will develop methods and skills to find relevant literature for his/her topic, to prepare methodologically sound scientific experiments, and to write a well-structured scientific paper and to present his/her results. He/she will be also aware of the need to avoid plagiarism.  The key qualification Scientific Research is highly recommended as a prerequisite for the seminar.

    Personal qualification: Students will acquire skills on how to find relevant literature for a research topic, organize a small research task, write a well-structured, concise paper about it and present the results of their work. He/she is well prepared to write and present a Master’s Thesis. 

  • Topics

    Dense distributed word representations (word embeddings) play a key role in the remarkable performance of deep learning models on challenging natural language processing tasks. While word embeddings are capable of capturing and condensing numerous semantic features of words, they are created by a black-box process and their numerical coordinates alone have no meaningful interpretation.

    Given the wide range of applications of language models and deep learning models in general, their interpretability is becoming increasingly important to increase trust in predictions, detect potential errors and biases or to comply with legal regulations.

    In this seminar we discuss different approaches to improve interpretability in the field of natural language processing. Individual topics and research articles will be assigned at the kick-off meeting.

  • Schedule

    The schedule below is preliminary, dates are subject to change.

    Registration period

    15.08.22 – 05.09.22 (11.59 pm) via Portal2
    Notification of acceptance07.09.22 
    Kick-off meeting


    10:15 am – 11:45 am

    assignment of seminar topics

    Drop-out until18.09.22 


    8.30 am – 6.15 pm

    midterm presentations


    8.30 am – 6.15 pm

    final presentations
    Submission deadlineto be announced 
  • Registration

    Registration is possible until Monday, 05.09.22 via Portal2.  There will be an allocation suitable to priorities on Tuesday, 07.09.22. After allocation you have the possibility to deregister until Sunday, 18.09.22.