让预测更准且可解释：结合决策规则的流程后续预测方法
Decision-Aware Suffix Prediction and Reasoning of Business Processes
- 融合决策挖掘规则与神经网络，实现可推理的预测
- 短前缀和罕见流程变体的预测准确率显著提升
- 适合需要可解释性与高精度的业务流程场景
后缀预测旨在推断运行中案例的剩余事件序列直至完成。现有方法多依赖事件日志训练的神经网络，虽平均表现良好，但在短前缀或罕见流程变体场景下表现不佳。此类情况下，正确路径常涉及多个由案例和事件级属性决定的分支决策，而神经网络模型往往因过度依赖密集事件标签而忽视这些决策信号。决策挖掘可从日志中提取决策规则，但以往仅用于事后分析和假设分析，未用于后缀预测。为此，本文提出一种决策感知的后缀预测框架，采用神经符号方法，使预测结果能通过挖掘出的决策规则进行推理。在四个事件日志中的三个及三种后缀预测器上的实验表明，该框架在短前缀和罕见变体上均提升预测性能，并带来内在可解释性。
原文摘要 · Abstract (English)
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 extracts rules for such decisions from the event log, but has so far been applied only to post-hoc and what-if analysis, not suffix prediction. We therefore extend suffix prediction with decision mining, introducing a decision-aware suffix prediction framework, a neuro-symbolic approach that enables reasoning about predicted events via mined decision rules. Experiments on three of four event logs and three suffix predictors show that the framework can improve suffix prediction, especially for short prefixes but also for rare process variants, and adds intrinsic interpretability.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。