Towards Reproducibility in Predictive Process Mining: SPICE - A Deep Learning Library
By: Oliver Stritzel , Nick Hühnerbein , Simon Rauch and more
In recent years, Predictive Process Mining (PPM) techniques based on artificial neural networks have evolved as a method for monitoring the future behavior of unfolding business processes and predicting Key Performance Indicators (KPIs). However, many PPM approaches often lack reproducibility, transparency in decision making, usability for incorporating novel datasets and benchmarking, making comparisons among different implementations very difficult. In this paper, we propose SPICE, a Python framework that reimplements three popular, existing baseline deep-learning-based methods for PPM in PyTorch, while designing a common base framework with rigorous configurability to enable reproducible and robust comparison of past and future modelling approaches. We compare SPICE to original reported metrics and with fair metrics on 11 datasets.
Similar Papers
On the Simplification of Neural Network Architectures for Predictive Process Monitoring
Machine Learning (CS)
Makes smart computer predictions much faster and smaller.
A Review of Artificial Intelligence Impacting Statistical Process Monitoring and Future Directions
Systems and Control
AI makes factories automatically fix problems.
Neuro-Symbolic Predictive Process Monitoring
Artificial Intelligence
Teaches computers to predict future steps logically.