← Back to all news
arXiv AI By Oliver Stritzel, Nick H\"uhnerbein, Simon Rauch, Itzel Zarate, Lukas Fleischmann, Moike Buck, Attila Lischka, Christian Frey

Towards Reproducibility in Predictive Process Mining: SPICE -- A Deep Learning Library

Read the original on arXiv AI →

The paper introduces SPICE, a Python framework that reimplements three popular deep‑learning methods for Predictive Process Mining (PPM) using PyTorch. It provides a common, highly configurable base to enable reproducible and robust comparison of PPM models, addressing issues of reproducibility, transparency, and usability. The authors benchmark SPICE against the original reported metrics and fair metrics across 11 datasets.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Machine Learning
Oct 21

Process-Aware Event Log Augmentation for Process Prediction under Labeled Data Scarcity

The paper introduces SiamSA-PPM, a self‑supervised framework that augments event logs for predictive process monitoring by generating realistic trace variants using control‑flow semantics and frequent behavioral patterns. It combines Siamese learning with statistically grounded data augmentation to learn generalizable prefix representations without requiring labeled data. Experiments on diverse real‑world logs show that SiamSA-PPM matches or outperforms state‑of‑the‑art baselines in next‑activity and final‑outcome prediction tasks.

By Sjoerd van Straten, Alessandro Padella, Marwan Hassani
benchmarks
More like this →