Decision-Aware Suffix Prediction and Reasoning of Business Processes
Quick summary
arXiv:2609.06169v1 Announce Type: cross Abstract: Suffix prediction forecasts the remaining sequence of events of a running case until completion. Most approaches rely on neural networks trained on event logs, which, on average, perform well but struggle with short prefixes or targets belonging to a rare process variant. In such scenarios, the correct path may cross multiple branching decisions, determined primarily by case- and event-level attributes, a signal that NN-based suffix prediction models tend to underweight because they may heavily weight (dense) event labels. Decision mining extra
Key takeaways
- arXiv:2609.06169v1 Announce Type: cross Abstract: Suffix prediction forecasts the remaining sequence of events of a running case until completion.
- Most approaches rely on neural networks trained on event logs, which, on average, perform well but struggle with short prefixes or targets belonging to a rare process variant.
- In such scenarios, the correct path may cross multiple branching decisions, determined primarily by case- and event-level attributes, a signal that NN-based suffix prediction models tend to underweight because they may heavily weight (dense) event labels.
Why it matters
“Decision-Aware Suffix Prediction and Reasoning of Business Processes” should be evaluated beyond branding and benchmark scores. Its practical importance will emerge in task accuracy, latency, unit cost, safety and integration with real workflows.
Member comments