After successful completion of the course, students are able to process and analyse data in the selected domain, select appropriate methods based on the requirements, apply methods to real data, and develop solutions for domain-specific tasks.

Project addressing a domain-specific challenge.

The interdisciplinary project is part of the corresponding module in the Data Science curriculum and forms a content block together with the lecture series 194.046 Interdisciplinary Lecture Series on Data Science, as well as the course 194.047 Domain-Specific Lectures in Data Science.

In principle, all topics/disciplines presented in the lecture series 194.046 Interdisciplinary Lecture Series on Data Science in previous years are available for the project, with the respective lecturers as main supervisors.

The domain-specific lecture, to be selected from the list available in TISS, provides the basic skills needed for the work in the corresponding interdisciplinary project. Therefore, that course should ideally be completed before or possibly during the work on this project.

Steps for the Interdisciplinary Project in Data Science

1. Select a main supervisor for the project

• Usually not from the Faculty of Informatics (or Mathematics)
• Not necessarily from the TU Wien
• All presenters of the associated Lecture Series (194.046) of the past years (as listed in the TUWEL course of the lecture series) are available to act as main supervisor for the project.

2. Discuss the project with the selected supervisor, agree on a 1 to 2-page proposal document and identify the corresponding domain-specific lecture in data science (from the list available for course 194.068)

3. Select a co-supervisor for the project

• Usually from the Faculty of Informatics or Mathematics
• Must be from the TU Wien
• Specifically, anybody that has lectured any of the Data Science courses (see list below)

4. Discuss the 1 to 2-page project proposal with the co-supervisor. Ensure, specifically, that the key methods covered inthe Data Science Curriculum, are referred to and will be properly applied in the project. (e.g. CRIS-DM, Experiment Design, Reproducibility, Data Science lecture specific methods)

5. Refine the project proposal until both supervisors agree and upload the final version on TUWEL

6. Do the project

7. Discuss regularly with the supervisors

8. Write the report  and prepare the poster

9. Once approved by both supervisors upload the final report and the poster to TUWEL

Note that a project with the company that you are currently working for is generally not a good fit to the requirements for this inter-disciplinary project.

Potential Main Supervisors

Look at the list of lecturers of the "Domain-Specific Lectures in Data Science" over the last years.

Potential Co-Supervisors

The following people are potential co-supervisors:

Alessio Arleo

Alexander Schindler

Allan Hanbury

Andreas Rauber

Cem Okulmus

Christian Bors

Davide Ceneda

Dimitrios Sacharidis

Elmar Kiesling

Fajar Ekaputra

Florina Piroi

Ivona Brandic

Jesper Larsson Träff

Klaus Nordhausen

Kresimir Matkovic

Manuela Waldner

Margit Pohl

Markus Zlabinger

Marta Sabou

Matthias Lanzinger

Nysret Musliu

Peter Filzmoser

Peter Knees

Rudolf Mayer

Sascha Hunold

Sebastian Hofstätter

Silvia Miksch

Theresia Gschwandtner

Tomasz Miksa

Victor Schetinger

Wolfgang Aigner

Solve a practical problem in inter-disciplinary project work

Write a report

Prepare a poster for presentation

Steps in the Module “DSA – Domain-Specific Aspects of Data Science”

1. Attend the Interdisciplinary Lecture Series on Data Science (194.046)
2. Choose an area
3. Get theoretical knowledge through attending a lecture in this area (3,0/2,0 VO/VU/SE Fachspezifische Lehrveranstaltungen)
4. Solve a practical problem in inter-disciplinary project work – Interdisciplinary Project in Data Science (194.060/194.047)

Proposal Outline:

• Title

• Your name and student ID number

• The name of the main supervisor (from the Lecture Series 194.046) and the co-supervisor (from lectures in the Data Science curriculum)

• Short abstract

• Motivation / problem statement / well-defined, non-trivial research question(s)

• The methodology and process being applied. Ensure, specifically, that the key methods covered in the Data Science Curriculum are referred to and will be properly applied in the project. (e.g. CRISP-DM, Experiment Design, Reproducibility, specific methods from other Data Science lectures)

• Expected results (KPI/Success Criteria) as well as approaches, baseline and metrics for evaluation

• Domain-specific lecture(s) selected (from the list available for course 194.068), plus indication whether already completed or to be done in parallel to the project

Report on the results

Final Certificate

If the supervisor and co-supervisor do not have the rights to issue a certificate, the supervisor or co-supervisor should send an e-mail to allan.hanbury@tuwien.ac.at (with all involved people in cc) containing the final grade for the project. The certificate will then be issued in the names of the supervisor and co-supervisor (as possible given their affiliation to the TU Wien).