M Maikel Leemans

Discovery of Frequent Episodes in Event Logs

Lion’s share of process mining research focuses on the discovery of end-to-end process models describing the characteristic behavior of observed cases. The notion of a process instance (i.e., the case) plays an important role in process mining. Pattern mining techniques (such as traditional episode mining, i.e., mining collections of partially ordered events) do not consider process instances. In this paper, we present a new technique (and corresponding implementation) that discovers frequently occurring episodes in event logs, thereby exploiting the fact that events are associated with cases. Hence, the work can be positioned in-between process mining and pattern mining. Episode Discovery has its applications in, amongst others, discovering local patterns in complex processes and conformance checking based on partial orders. We also discover episode rules to predict behavior and discover correlated behaviors in processes, and apply our technique to other perspectives present in event logs. We have developed a ProM plug-in that exploits efficient algorithms for the discovery of frequent episodes and episode rules. Experimental results based on real-life event logs demonstrate the feasibility and usefulness of the approach.

Process mining in software systems: Discovering real-life business transactions and process models from distributed systems

This paper presents a novel reverse engineering technique for obtaining real-life event logs from distributed systems. This allows us to analyze the operational processes of software systems under real-life conditions, and use process mining techniques to obtain precise and formal models. Hence, the work can be positioned in-between reverse engineering and process mining. We present a formal definition, implementation and an instrumentation strategy based the joinpoint-pointcut model. Two case studies are used to evaluate our approach. These concrete examples demonstrate the feasibility and usefulness of our approach.

Modeling and Discovering Cancelation Behavior

This paper presents a novel extension to the process tree model to support cancelation behavior, and proposes a novel process discovery technique to discover sound, fitting models with cancelation features. The proposed discovery technique relies on a generic error oracle function, and allows us to discover complex combinations of multiple, possibly nested cancelation regions based on observed behavior. An implementation of the proposed approach is available as a ProM plugin. Experimental results based on real-life event logs demonstrate the feasibility and usefulness of the approach.

Hierarchical performance analysis for process mining

Process mining techniques use event data from operational and software processes to discover process models, to check the conformance of predefined process models, and to extend such models with information about bottlenecks, decisions, and resource usage. In recent years, the process mining field made huge advances in terms of scalability. In addition, recent work in process discovery supports advanced process model constructs such as subprocesses, recursive structures, cancellation, and various notions of concurrency. Hence, one has to realize that a simple, small, and flat model will not suffice anymore, especially when applied to analyzing software system processes. However, state of the art performance analysis is still typically performed either over the whole process model or at the level of individual activities. There is a lack of formal support for performance analysis on various submodel abstractions while taking into account the execution semantics. This paper presents 1) a framework for establishing precise relationships between events and submodels, taking into account execution semantics; and 2) a novel formalization of existing and novel performance metrics. Our approach enables advanced performance analysis at various submodel abstractions. An implementation is made available, and we demonstrate the advantages of our approach to various software system processes, showing the applicability and advantage with respect to existing techniques.