TracePrism helps find distinct sub-processes in event logs iteratively
TracePrism: Extending Semantic Trace Topics to Iterative, Encoding-Aware Trace Partitioning
Human-Computer Interaction
Summary
When companies try to understand their workflows from event data, they often create one big model for all processes, which can miss important differences. The authors offer TracePrism, a tool that helps find separate groups of similar processes step-by-step, using different ways to describe each group. Analysts can revise their groupings as they learn more, making it easier to uncover hidden process patterns and changes over time. They tested TracePrism on real data and found it reveals unexpected behaviors like process changes.
What this means in practice
- •For business process analysts: Separate mixed workflow cases in company event logs into meaningful groups to improve understanding and modeling of distinct sub-processes.
- •For process mining software developers: Build adaptive process mining tools that support iterative selection and revision of trace encodings and groupings to better capture evolving workflows.
Authors
Gennady Andrienko, Natalia Andrienko, Manuel Resinas, Stef van den Elzen, Barbara Weber
Abstract
Process mining typically explains an event log with a single model, clustering, or classifier fitted to all cases, although real logs often mix cases that follow different sub-processes. Trace clustering separates sub-populations, but it fixes one notion of trace similarity for the whole analysis. We propose TracePrism, which treats process discovery as iterative partitioning with three characteristics. The trace encoding is adaptive: it is chosen for each working set of remaining traces from a library of encodings, with suggestions scored for that set. The partitioning is iterative: the analyst forms a candidate group based on a rule, a dominant topic, a cluster over topic weights, or a region of an embedding, accepts it if it is coherent in both process structure and outcome, and continues with the remaining traces. Decisions are revisable and recorded: a group can be reopened under a different working set or encoding, and the findings repository keeps both the earlier decision and its revision. We implement TracePrism as an open-source prototype and demonstrate it in usage scenarios on three real-world event logs. The scenarios also revealed behavior that we had not anticipated, such as process drift.