Instructors: Dr. Manfred Eppe
Event type:
General Professional Skills courses
Displayed in timetable as:
MKKI-basiert Proz
Hours per week:
2
Language of instruction:
German
Min. | Max. participants:
- | 30
Registration group: Anmeldegruppe Methodenkompetenz
Comments/contents:
How can we draw conclusions and test hypotheses from the ever-increasing amounts of data generated by scientific experiments? How can these data be visualized and analyzed? Which tools are useful for this and how can one use them?
This course addresses these questions and deals with modern AI-based methods of data science. Students will learn to use the Python programming language and data science libraries to formulate hypotheses and to verify or falsify them on the basis of collected data. In particular, artificial neural networks are used for classification, regression and grouping of data.
Besides the practical aspects, a great deal of attention is paid to scientific work. What is a research question, and how is it formulated? What is a research goal? How do you write a research paper, and how do you do it all as a team? These questions are dealt with in a mini-project.
Prior knowledge of Python, statistics and other basics is advantageous but not required. More important is a personal enthusiasm for scientific work and artificial intelligence.
Learning objectives:
The seminar serves to convey the following competencies:
* Using and understanding AI-based data science tools
* Creating and formulating research questions and hypotheses
* Presenting and evaluating a complex topic in a comprehensible way in a paper
* Presenting a complex topic in an understandable way with slides
* Scientific work in a team
In addition, the corresponding technical contents are to be learned within the scope of a mini-project.
Didactic concept:
The course will be held digitally. The course will be realized with Zoom and learning videos via Lecture2Go. It is generally necessary that the students are virtually/digitally present during the course times.
The first part will deal with classical methods of data science. Thereafter, the students will start generating preliminary results for their mini-projects. In the second part, neural networks will be used, and the previous results will be compared with the results of the neural networks. The different methods can generate different results for the same research question. This trains the students to discuss the methods and to draw conclusions about the different results. This concluding discussion will be presented during the final presentation.
Literature:
* Géron, A. (2017). Hands-on machine learning with Scikit-Learn and TensorFlow: concepts, tools, and techniques to build intelligent systems. O’Reilly Media.
* VanderPlas, J. (2016). Python Data Science Handbook: essential tools for working with data. O’Reilly Media.
* www.kaggle.com
Additional examination information:
The grade will be based on the final presentation and report.
|