Directed acyclic graph extraction from event logs

Abstract

The usage of probabilistic models in business process mining enables analysis of business processes in a more efficient manner. Although, the Bayes-ian belief network is one of the most common probabilistic models, possibilities to use it in business process mining are still not widely researched. Existing process mining approaches are incapable to extract directed acyclic graphs for representing Bayesian networks. This paper presents an approach for extraction of directed acyclic graph from event logs. The results obtained during the ex-periment show that the proposed approach is feasible and may be applied in practice.