Education

Resources

• Formal Requirements and Procedure: Describes the formal requirements for your work (e.g., language and length), recommended tools, and the complete process from proposal writing and seminar attendance to submission and grading.
• How to Prepare a Proposal: Describes the key elements that your proposal must contain.
• From Research to Thesis: Principles of Scientific Writing and Argumentation: Provides guidelines for scientific writing and support for structuring a thesis.
• Giving a Presentation — Checklist for the PRG Seminar: Provides guidelines for giving a presentation in our seminar.
• Sample Templates
  • BSc template
  • MSc template
  • Presentation template

Available projects for BSc and MSc theses

Below is a list of possible topics for BSc and MSc theses in the area of pattern recognition and related fields. In addition, many further projects in pattern recognition and machine learning are possible. If you have an interesting idea or would like to develop a project together with us, please do not hesitate to contact a team member.

We also offer topics for BSc theses in the area of human–computer interaction. If you are interested in this field, please contact us as well.

The detailed scope of each project is defined in consultation with the supervisor. In particular, the listed projects can be carried out either as a BSc or an MSc thesis, depending on the agreed scope.

Title	Contact
Automated Metadata Extraction and Reconstruction of Case Progression in Swiss Court Decisions	kaspar.riesen@unibe.ch
Few Shot Learning on the Swiss River Network	benjamin.fankhauser@unibe.ch
Supervised Learning models to analyze X-Ray Diffraction data	kaspar.riesen@unibe.ch
Data Exploration & Visualization for High-Dimensional Material Property Maps	kaspar.riesen@unibe.ch
Predicting Molecular Energy from QTAIM-derived Multi-granularity Descriptors using Graph Neural Networks and Transformers	linlin.jia@unibe.ch
Modeling Dynamic Intraoperative Physiological Signals via Spatio-temporal Graphs for Predicting Postoperative Adverse Outcomes	linlin.jia@unibe.ch
DeepSchemata: Towards Unified Graph Extraction for Technical Drawings	linlin.jia@unibe.ch
Are Vision Large Language Models Effective Handwriting Graph Extractors?	linlin.jia@unibe.ch
Tangle-Based Heuristics for Subgraph Matching in Large Networks	kalvin.dobler@unibe.ch
GAUGE: Graph Attention with Universal and Global Extensions	merlin.streilein@unibe.ch
SCOPE: Summarization with Context-Oriented Preprocessing and Evaluation	merlin.streilein@unibe.ch
Reinforcement Learning for Continuous PV Power Forecasting with Reward-Based Model Updates	fereshteh.jafari@unibe.ch
Transfer Learning and Domain Adaptation for PV Prediction Across Different Geographical Locations	fereshteh.jafari@unibe.ch
Adaptive Online Learning for Short-Term PV Power Prediction	fereshteh.jafari@unibe.ch
Smart Home Energy Management App with PV Prediction	fereshteh.jafari@unibe.ch
Spatiotemporal Gait Parameters from Wrist-Worn Accelerometers	aaron.colombo@unibe.ch
Analysis of controlled augmentation in SE(3) environments	francesco.leonardi@unibe.ch
Learning and analysis of rotations in SE(3) environments	francesco.leonardi@unibe.ch
Solving Real World Problems with Pattern Classification	kaspar.riesen@unibe.ch

Completed Thesis Projects

Courses