Discovering behavioural patterns in knowledge-intensive collaborative processes

Domains like emergency management, health care, or research and innovation development, are characterized by the execution of so-called knowledge-intensive processes. Such processes are typically highly uncertain, with little or no structure; consequently, classical process discovery techniques, aimed at extracting complete process schemas from execution logs, usually provide a limited support in analysing these processes. As a remedy, in the present work we propose a methodology aimed at extracting relevant subprocesses, representing meaningful collaboration behavioural patterns. We consider a real case study regarding the development of research activities, to test the approach and compare its results with the outcome of classical process discovery techniques.