We set up collaboration projects with our partners from the private sector. These allow students to directly put the skills acquired during class into action, gain hands-on experience in a professional data science setting, and receive feedback and mentoring from seasoned data scientists. We announce collaboration projects in class and also discuss there how to apply.
What is the project about?
The project is about finding (and applying) methods to identify anomalies and abnormal observations in production processes. Provided with datasets by Daimler AG, the task of the student is to explore the datasets and their respective structures and based on these insights, choose appropriate methods to conduct anomaly analysis, i.e., find (groups of) observations that differ significantly from the normal case.
What is the scope of the project?
The main focus of this project is to compare different methods and to discuss the advantages and disadvantages with respect to their theoretical assumptions and their practical implications (e.g., computational costs). In this context, it is noteworthy that Daimler AG is interested in applying the methods provided by the student to different datasets after the project has been submitted. Consequently, reproducibility and setting up a flexible data workflow is a core requirement of the project. While guiding literature is provided, the choice of the methods is up to the student. Therefore, methods applied can range from traditional econometric time series analysis to the application of machine- and deep learning techniques.
Why should I be interested?
Overall, the project enables the student to learn about the use cases of data analysis in a corporate environment and get hands-on experience in professional data science. Through supervising meetings with Daimler AG the student has additionally the chance to get feedback and insights from professional data scientists and learn about their work.