|M.Sc Student||Yosef Dahari|
|Subject||Model Synthesis in Process Mining|
|Department||Department of Industrial Engineering and Management||Supervisor||Full Professor Gal Avigdor|
|Full Thesis text|
Modern information systems record the execution of transactions as part of business processes in event logs. This thesis relates to process mining techniques that aim at discovering process models based on event-logs namely, process discovery. Recently, various discovery algorithms have been proposed, each with specific advantages and limitations. Targeting the improvement of process discovery, we argue that, instead of relying on a single algorithm, the outcomes of different discovery algorithms shall be fused to combine the strengths of individual approaches. We propose a general framework for such fusion and instantiate it with two new discovery algorithms: The Exhaustive Noise-aware Inductive Miner (exNoise), which, exhaustively searches for model improvements; and the Adaptive Noiseaware Inductive Miner (adaNoise), which is a computationally tractable version of exNoise. For both algorithms, we formally show that they outperform each of the individual mining algorithms used by them. Our empirical evaluation further illustrates that fusion-based discovery yields models of better overall quality than state-of-the-art approaches, and that the theoretical guarantees holds in practice. For evaluating process models we considered multiple quality dimensions to show that fusion-based models are more balanced and precise. This framework can be further instantiated to fuse discovery algorithms, in a quality-aware fashion.