Answers are grounded in this paper and cite their sources.
Semi-structured processes are data-driven, human-centric, flexible processes whose execution between instances can vary dramatically. Due to their unpredictability and data-driven nature, it's becoming increasingly important to mine traces of events collected from these processes. This enables the extraction of mined process models that could help users handle new process instances. Process-mining techniques can help facilitate this goal, but it can be daunting for users new to process-aware analytics to sift through the literature and available software to determine which process-mining algorithm to use. The authors compare five process-mining algorithms and present a decision tree to help readers determine which mining algorithm to use for a specific problem. Semi-structured processes, however, present challenges that these mining techniques don't address. So, the authors also identify three key characteristics of semi-structured processes and the mining challenges they present, highlighting a selection of emerging mining approaches that can address these challenges.
No takes yet. Share an insight, caveat, or question.
Lakshmanan et al. (2012) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: