{"ID":2837379,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.18830","arxiv_id":"2511.18830","title":"Leveraging Duration Pseudo-Embeddings in Multilevel LSTM and GCN Hypermodels for Outcome-Oriented PPM","abstract":"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.","short_abstract":"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 at...","url_abs":"https://arxiv.org/abs/2511.18830","url_pdf":"https://arxiv.org/pdf/2511.18830v1","authors":"[\"Fang Wang\",\"Paolo Ceravolo\",\"Ernesto Damiani\"]","published":"2025-11-24T07:06:08Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[]","has_code":false}