Bayesian belief network application in process mining

Abstract

Business process mining gains more and more attention from both scientific and business communities. Its value has been recognized in information systems reverse engineering field as it allows discover knowledge about the implementation and execution of business processes in order to analyse and improve them. The probabilistic model of business process execution is essential for its analysis,however, there is very little research done on this topic. In this paper, we present a novel approach on automatic extraction of probabilistic models from mined business processes. We present an example of application of our approach in a real world scenario. The results achieved by application of our approach can faciliate decision making for the business users and therefore may be used for process analysis or enhancement of information systems with additional decision support capabilities.