Knowledge Representation and Machine Learning

In summer term, we offer the courses:

• User Studies in interactive intelligent Systems (Seminar + Exercises) for bachelor students in the module 39-Inf-WP-MTI (Human-Machine Interaction) and masters students in the modules 39-M-Inf-INT-app Applied Interaction Technology or 39-M-Inf-INT-app-foc Applied Interaction Technology (focus).
• Human-Centric Machine Learning (Seminar + Exercises) for bachelor students as part of the module 39-Inf-WP-IG Computer Science and Society.

In winter term, we offer the courses:

• Introduction to Data Mining (lecture + exercises) for bachelor students in the module 39-Inf-DM Grundlagen Datamining, 39-Inf-WP-DS Data Science (Basis) and 39-Inf-WP-DS-x Data Science (Schwerpunkt), as well as for master students in the module 39-M-Inf-AI-bas Basics of Artificial Intelligence.
• Applied Optimization (lecture + exercises) as mandatory course for data science masters students; and as voluntary course for bachelor students in the module 39-Inf-DM Grundlagen Datamining, 39-Inf-WP-DS Data Science (Basis) and 39-Inf-WP-DS-x Data Science (Schwerpunkt), as well as for master students in the module 39-M-Inf-AI-bas Basics of Artificial Intelligence.

We are also active in further activities to improve teaching with our research.

Theses

Currently, our capacities to supervise theses are limited and all our open topics have already been taken. Still, in special cases, like emergencies where a thesis topic needs to be found on short notice, or topics that are so specialized that only we can supervise them, please contact us directly:

Prof. Benjamin Paassen
+49 521 106-87838
bpaassen@techfak.uni-bielefeld.de

Theses in our group are typically in the areas:

• Educational data mining, learning analytics and machine learning for education
• Intelligent tutoring systems
• Machine learning for structured data (e.g. graph neural networks, autoencoders for structures, edit distances)
• Machine learning with prior knowledge and data-sparing machine learning (e.g. few-shot learning, transfer learning)
• Interpretable and explainable machine learning
• Fairness in machine learning

The following theses have already been completed in our group

• Development and Evaluation of a Categorization Interface for Social Psychology Experiments (MA thesis by Julian Goergen; supervised by Prof. Benjamin Paaßen and Prof. Thekla Morgenroth)
• Adaptive Lernumgebungen basierend auf Cognitive Load - Evaluation anhand eines Prototyps zum Thema Wald (MA thesis by Nikolas Weber; supervised by Prof. Benjamin Paaßen and Dr. Daniela Sellmann-Risse)
• Prototype-Based Convolutional Neural Networks for EMG signal classification (BA thesis by Dennis Schielke; supervised by Prof. Benjamin Paaßen and Rui Liu)
• Design and Evaluation of a Word-Clustering-Tool for Social Psychology (BA thesis by Luis Klocke; supervised by Prof. Benjamin Paaßen and Dr. Adia Khalid)
• Partial Credit Item Response Autoencoders (BA thesis by Lukas Wüppelmann; supervised by Prof. Benjamin Paaßen and Jesper Dannath)
• Prototype Decision Trees via Learning Vector Quantization (BA thesis by Arno Gaußelmann; supervised by Prof. Benjamin Paaßen and Valerie Vaquet)
• Automatic Unit-Tests Generation via Large Language Models for Feedback in Programming Education (BA thesis by Lukas Morasch; supervised by Prof. Benjamin Paaßen and Jesper Dannath)
• Comparing Large Language Models for Automatic Unit Test Generation in a Python Course (BA thesis by Richard Pamies; supervised by Prof. Benjamin Paaßen and Jesper Dannath)
• Text-based difficulty estimation for multiple choice questions (BA thesis by Niklas Rotter; supervised by Prof. Benjamin Paaßen and Alonso Palomino Garibay)
• Comparing Two-State Hidden Markov Models with Covariates to a Hierarchical approach using Clustering and Logistic Regression (BA thesis by Özay Öztürk; supervised by Prof. Benjamin Paaßen and Prof. Roland Langrock)
• Few-Shot Multilabel Text Classification of Student Counseling Questions (BA thesis by Jasper Matzat; supervised by Prof. Benjamin Paaßen and Jesper Dannath)
• Personalized Chess Training Plans via Multidimensional Item Response Theory (BA thesis by Luca Sander; supervised by Prof. Benjamin Paaßen and Alina Deriyeva)
• Generative KI im agilen Software Engineering - Eine Fallstudie in der universitären Ausbildung (BA thesis by Luca Strignano; supervised by Prof. Benjamin Paaßen and Dr. Sebastian Wrede)
• Adapt-Leaf-Forest: ein innovativer Ansatz für die Klassifikation mit Random Forest (BA thesis by Maurice Heidemann; supervised by Prof. Benjamin Paaßen and Prof. Kevin Tierney)
• Learning to Style Check (MA thesis by Tim Marvin Heptner; supervised by Prof. Benjamin Paaßen and Dr. Alexander Schulz)
• Optimal Team Composition and Skill Estimation in Pair Sports (MA thesis by Sören Rüttgers; supervised by Prof. Benjamin Paaßen and Dr. Ulrike Kuhl)