Decision mining
Decision mining is the process-mining extension that discovers the data-driven rules behind branching decisions in a process — why this case took variant A and that one took variant B. Decision rules are extracted from case attributes using machine-learning classifiers (decision trees most commonly) and presented as interpretable conditions.
Decision mining matters because process mining alone tells you 'three variants exist'; decision mining tells you 'variant A happens when customer tier = enterprise AND order value > $10K'. That conversion from process diversity to explanatory rules is what makes process-mining output actionable: now you can ask whether the rule makes sense, whether it should be codified into the process, or whether the variant exists because of a defect in how the case attributes were captured. The discipline is closest to causal inference; the output is most useful when reviewed by domain experts who can validate the rules.
Related terms
- Variant analysis
Variant analysis groups cases by the unique sequence of activities they followed and ranks variants by frequency and cost.
- Process discovery
Process discovery is the process-mining technique of constructing a process model (typically BPMN or a Petri net) from an event log without prior knowledge of the intended process.
- Process conformance
Process conformance is the analysis of how well an observed event log matches a reference process model — measuring deviations, missing steps, and extra steps.