Discovering high-level process models from event logs

Process mining is a relatively new field of computer science, which deals with process discovery and analysis based on event logs. In this paper we consider the problem of discovering a high-level business process model from a low-level event log, i.e. automatic synthesis of process models based on the information stored in event logs of information systems. Events in a high-level model are abstract events, which can be refined to low-level subprocesses, whose behavior is recorded in event logs. Models synthesis is intensively studied in the frame of process mining research, but only models and event logs of the same granularity are mainly considered in the literature. Here we present an algorithm for discovering high-level acyclic process models from event logs and some specified partition of low-level events into subsets associated with abstract events in a high-level model.