Doctoral meeting: 'Structural change detection on process models'
Process Mining aims to discover, monitor and improve real processes by extracting information from execution logs readly available in today's information systems. It includes tasks as model discovery from executions, quality checking for that models and enhacement of the running processes with the knowledge that can be extracted from real cases. Real life processes, that are executed in dynamic environments, change along time. In this context, it is important to detect this changes to prevent organizations from undesired outcomes or productivity drops.
In this thesis we propose the creation of some techniques that are able to detect these changes with a short delay. To do that we use conformance metrics and discovery algorithms, which are evaluated and monitored over time in order to detect when a process model changes. With this approach we are able to detect changes with a high accuracy, overcoming the problems present in the approaches from the state of the art.