Process Discovery using Inductive Miner and Decomposition
This work addresses the challenge of process discovery for business process management, but it is incremental as it applies existing methods to a contest setting.
The authors tackled the problem of discovering business process models from event logs by using Inductive Miner and Decomposition techniques to balance overfitting and underfitting, achieving results through classification tasks on training and test sets with validation and complexity analysis.
This report presents a submission to the Process Discovery Contest. The contest is dedicated to the assessment of tools and techniques that discover business process models from event logs. The objective is to compare the efficiency of techniques to discover process models that provide a proper balance between "overfitting" and "underfitting". In the context of the Process Discovery Contest, process discovery is turned into a classification task with a training set and a test set; where a process model needs to decide whether traces are fitting or not. In this report, we first show how we use two discovery techniques, namely: Inductive Miner and Decomposition, to discover process models from the training set using ProM tool. Second, we show how we use replay results to 1) check the rediscoverability of models, and to 2) classify unseen traces (in test logs) as fitting or not. Then, we discuss the classification results of validation logs, the complexity of discovered models, and their impact on the selection of models for submission. The report ends with the pictures of the submitted process models.