The REVisPro Project sits at the intersection of process compliance and graph visualization with focus on representing and visualizing rules extracted from text.
Ensuring business processes comply with evolving regulatory and legal frameworks remains a highly resource-intensive and error-prone task. Standard compliance verification requires explicit modeling of workflows alongside the extraction, formalization, and visualization of rules from dense text. For small and midsize enterprises (SMEs), executing this pipeline is frequently unfeasible due to a lack of specialized technical and legal expertise. Consequently, process managers struggle to map relevant policies to dynamic workflows, while compliance managers often identify violations reactively during costly audits.
To address these challenges, the interdisciplinary REVisPro project aims to create generative AI-driven methods to automate and simplify process compliance management for non-technical users.
Are you a small or medium enterprise? If you are interested in working with us on your own compliance rules, regulations and documents, you can contact us via ReVisPro@bpm.cit.tum.de
Core Research Goals
- GenAI-Driven Extraction: Leveraging generative AI to automatically extract compliance rules from complex legal texts.
- Traceable Visualizations: Designing visual interfaces that interlink extracted rules, operational workflows, and the original source texts.
- Graph Analytics: Devising efficient algorithms and visualization techniques to support the interactive exploration of complex compliance rule graphs.
ReVisPro is a HDSC Tandem Project conducted in collaboration with Prof Stephen Kobourov from the chair of efficient algorithms.
