Finding suitable activity clusters for decomposed process discovery

Event data can be found in any information system and provide the starting point for a range of process mining techniques. The widespread availability of large amounts of event data also creates new challenges. Existing process mining techniques are often unable to handle "big event data" adequately. Decomposed process mining aims to solve this problem by decomposing the process mining problem into many smaller problems which can be solved in less time, using less resources, or even in parallel. Many decomposed process mining techniques have been proposed in literature. Analysis shows that even though the decomposition step takes a relatively small amount of time, it is of key importance in finding a high-quality process model and for the computation time required to discover the individual parts. Currently there is no way to assess the quality of a decomposition beforehand. We define three quality notions that can be used to assess a decomposition, before using it to discover a model or check conformance with. We then propose a decomposition approach that uses these notions and is able to find a high-quality decomposition in little time. Keywords: decomposed process mining, decomposed process discovery, distributed computing, event log