Pattern Recognition Group

About us

The Pattern Recognition Group (PRG) conducts research on algorithms and complex data structures in the fields of pattern recognition, machine learning, and data science. The group’s expertise is built on two complementary pillars:

• Graph-based representations for intelligent information processing.
• Document analysis using novel methods from natural language processing (NLP) and large language models (LLMs).

With the rapid advancement of storage technologies and data acquisition methods, the volume of available data in science and industry is growing at an unprecedented pace. At the same time, much of this data is inherently complex, making graphs one of the most powerful and suitable data structures for modeling and analysis. In parallel, the increasing importance of unstructured text and documents requires advanced language technologies to unlock and process the knowledge contained within.

These developments highlight the need for robust and efficient methods that help humans to interpret and manage large, complex datasets. The PRG addresses this challenge by developing novel graph-based algorithms for pattern recognition and data science as well as innovative document analysis techniques leveraging NLP and LLMs—always with the goal of providing feasible and reliable solutions that advance both theory and practice.

The Pattern Recognition Group (PRG) has been established in 2020 by Kaspar Riesen at the University of Bern.