Current research team

Bart Baesens	Supervisor
Jochen De Weerdt	PhD

Description Process mining

The topic of process mining is relatively new and can be situated at the intersection of the fields of Business Process Management (BPM) and data mining. It is inherently related to data mining and to the more general domain of knowledge discovery in databases (KDD) since the nature of its objectives is extracting useful information from large data repositories. Likewise, process mining is strongly associated with BPM because of its purpose of gaining insight into business processes. As a result, process mining fits flawlessly into the BPM life cycle framework. An illustration of process mining is shown in the figure below.


It should be noted that business process mining comprises process discovery because process mining describes a family of a-posteriori analysis techniques for extracting knowledge from event logs while process discovery only deals with extracting control-flow models. However, most of the attention in the process mining literature has been given to process discovery techniques.


Recent years have witnessed the ability to gather an enormous amount of data in a large number of domains. These data, for the most part, remain in repositories where they are never used. There is an urgent need to beneficially use these data to retrieve useable knowledge. Process discovery from event logs is one such attempt to generate actionable knowledge. This field of study is part of process mining research, which can be defined as the gathering of useful knowledge from information system audit trails or event logs. However, process discovery's main objective is the extraction of control-flow models from event logs.


Organizations currently face an information paradox: the more they automate their processes, the less they are capable of monitoring and understanding them. A good understanding of processes is nonetheless vital for fulfilling business requirements such as verifying and guaranteeing business process compliance, setting up a coherent access control policy and optimizing and redesigning business processes. A better understanding will eventually enable organizations to provide better, automated support for their business processes in flexible, process-aware information systems. Traditionally, practitioners have been obtaining insight into processes using interviewing techniques. However, factors like extensive business process automation, strategic behavior with interviewees and socio-economic factors such as personnel attrition, mergers and acquisitions have often shown these inherently qualitative interviewing techniques to be less suitable. A new and promising way of acquiring insights into business processes is the analysis of the event logs of information systems. In many organizations, such event logs conceal an untapped reservoir of knowledge about the way employees and customers conduct every-day business transactions. Event logs are already available in many organizations. Popular Enterprise Resource Planning (ERP) systems such as SAP R/3, Oracle e-Business Suite and workflow management systems (WfMS) such as ARIS, TIBCO and Microsoft Biztalk already keep track of these event logs.
The research team of professor Baesens currently studies the following topics in this context:

• The use of artificially generated negative events for the discovery of process models from event logs
• Application of machine learning techniques and probabilistic learning approaches to the analysis of event logs
• Assessment of currently available process discovery evaluation metrics and development of new process discovery metrics
• Industry applications of process mining: telecom industry, banking industry, government

Past research