Process Mining towards Semantics

Process mining techniques target the automatic discovery of information about process models in organizations. The discovery is based on the execution data registered in event logs. Current techniques support a variety of practical analysis, but they are somewhat limited because the labels in the log are not linked to any concepts. Thus, in this chapter we show how the analysis provided by current techniques can be improved by including semantic data in event logs. Our explanation is divided into two main parts. The first part illustrates the power of current process mining techniques by showing how to use the open source process mining tool ProM to answer concrete questions that managers typically have about business processes. The second part utilizes usage scenarios to motivate how process mining techniques could benefit from semantic annotated event logs and defines a concrete semantic log format for ProM. The ProM tool is available at www.processmining.org.