Introduction to Data Science: Processes, Analysis & Ethics
Two-Day Executive Course: Responsible Data Science in Practice
This two-day executive course provides professionals from various disciplines with a solid foundation in the responsible use of data science techniques. Using the CRISP-DM model (for a brief explanation, see this article on Medium), and tools such as the Ethical Data Assistant (DEDA) and the Human Rights and Algorithm Impact Assessment (IAMA), participants are challenged to approach complex data analysis with an eye for ethics, uncertainty, and societal impact. Not a standard step-by-step guide, but a practical toolkit for critical decision-making in data science and AI projects.
What does this program offer?
Participants will learn to:
- understand ethical risks and societal impacts of data science
- use CRISP-DM as a framework for the full development lifecycle of AI
- reflect on the role of uncertainty, dark data, and false assumptions
- translate scientific insights into practical applications within their own organization
- select techniques suited to specific data challenges
- analyze real-world cases using tools such as IAMA and DEDA
Who is this course for?
This course is designed for managers, developers, policy makers, and (data) scientists who want to learn how to apply data science responsibly.
Core Program Components
Program Component | Description |
|---|---|
Data Ethics I | Introduction to ethical considerations using DEDA and IAMA |
The Anatomy of an Answer | Dealing with uncertainty, missing data, and model validity |
CRISP-DM and the Cyclical AI Process | Structure within the data science process |
Connection to Practice | Reflection on application within one’s own organization |
Challenges in Data: Techniques | Exploration of tailored current analytical techniques (e.g., time series, missing data) |
Elective Modules (select 2 or 3)
Program Component | Description |
|---|---|
Data Curation & Wrangling | From raw data to usable datasets |
Data Ethics II | Applying IAMA to your own case |
Concept Drift & Non-Stationarity | Techniques to integrate the time dimension in data science |
Dark Data & Missingness | Recognizing and handling incomplete data |
Digital Twins | Technical and ethical implications of digital twins |
Monitoring ML & Evaluation | Oversight of algorithmic decision-making |
Practical information
- Duration: 2 days (consecutive or spaced apart)
- Location: Utrecht University or in-house
- Format: Modular and tailored to needs
- Instructors: Iris Muis, Gerko Vink, Stef van Buuren, Georg Krempl, Hakim Qahtan, Laurence Frank
Contact and more information:
Contact for questions Laurence Frank, Coordinator of Professional Education in Data Science & AI.
This course was created through a collaboration between the Department of Computer Science at the Faculty of Science, Data School at the Faculty of Humanities and Methods and Statistics at the Faculty of Social Sciences.