用时间伪嵌入提升流程预测模型的泛化能力
Leveraging Duration Pseudo-Embeddings in Multilevel LSTM and GCN Hypermodels for Outcome-Oriented PPM
- 分离事件与序列属性，用时长伪嵌入编码时间信息
- 在多种数据集上提升预测准确率，降低模型复杂度
- 适合需要时间敏感建模的工业流程监控场景
现有预测流程监控（PPM）的深度学习模型难以应对时间不规则性，尤其是随机事件时长和重叠时间戳，限制了其在异构数据集上的适应性。本文提出一种双输入神经网络策略，将事件与序列属性分离，利用时长感知的伪嵌入矩阵将时间重要性转化为紧凑可学习的表示。该设计应用于两类基线模型：B-LSTM 和 B-GCN，及其时长感知变体 D-LSTM 与 D-GCN。所有模型均集成自调参超模型以实现架构自适应选择。在平衡与非平衡结果预测任务上的实验表明，时长伪嵌入输入能持续提升模型泛化能力，减少模型复杂度，并增强可解释性。结果验证了显式时间编码的优势，为鲁棒、真实世界的 PPM 应用提供了灵活设计。
原文摘要 · Abstract (English)
Existing deep learning models for Predictive Process Monitoring (PPM) struggle with temporal irregularities, particularly stochastic event durations and overlapping timestamps, limiting their adaptability across heterogeneous datasets. We propose a dual input neural network strategy that separates event and sequence attributes, using a duration-aware pseudo-embedding matrix to transform temporal importance into compact, learnable representations. This design is implemented across two baseline families: B-LSTM and B-GCN, and their duration-aware variants D-LSTM and D-GCN. All models incorporate self-tuned hypermodels for adaptive architecture selection. Experiments on balanced and imbalanced outcome prediction tasks show that duration pseudo-embedding inputs consistently improve generalization, reduce model complexity, and enhance interpretability. Our results demonstrate the benefits of explicit temporal encoding and provide a flexible design for robust, real-world PPM applications.
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。