Research
Currently, our main research areas are:
Also check our web-sites on current and past research projects and our recent publications.
Software evolution and change impact analysis
In this area, we investigate techniques and tools to precisely determine the impacts of changes within and across software systems, ideally before they are committed to the code repository. Regarding the changes, we focus on source code and configuration changes and consider different programming paradigms and domains of software systems, such as object-oriented, component-based, and service-oriented software systems. Regarding our techniques, we explore the combination of static and formal program analysis (e.g., symbolic execution) with machine learning, in particular large language models (LLMs), to determine the impact.
Automating software engineering tasks
In this large research area, we exploit the repetitiveness in source code to investigate classical machine and deep learning methods (e.g., LLMs) to develop new methods, techniques, and tools to automate various software engineering tasks. We often combine these probabilistic methods with static program analysis and formal methods (e.g., symbolic execution) to obtain more deterministic and robust solutions. Regarding the engineering tasks, we currently focus on:
- Detecting and fixing bugs and vulnerabilities in programs
- Detecting and fixing errors in build configurations and/or full CI/CD pipelines
- Creating, augmenting, and selecting test cases
- Documenting source code and code changes
Quicklinks
Portals
Information for
Address
Universitätsstraße 65-67
9020 Klagenfurt am Wörthersee
Austria
+43 463 2700
uni [at] aau [dot] at
www.aau.at
Campus Plan
