A new benchmarking study published on arXiv investigates the effectiveness of various methods for predicting process-level behavior, a task distinct from typical trace-level predictions. The research compares statistical and machine-learning approaches, foundation models, and business process simulation against established trace-level predictive process monitoring (PPM) techniques. Results indicate significant performance variations among the methods, with aggregated trace-level predictions showing a tendency to compound errors at the process level, underscoring the need for dedicated research into process-level prediction. AI
IMPACT Highlights the need for dedicated research into process-level prediction, potentially influencing future development of AI tools for business process management.
RANK_REASON The cluster contains an academic paper detailing a benchmarking study of predictive methods. [lever_c_demoted from research: ic=1 ai=0.7]
- alphaXiv
- arXiv
- Business Process Simulation for the Design of Sustainable Product Service Systems (PSS)
- CatalyzeX Code Finder for Papers
- CORE Recommender
- DagsHub
- foundation model
- Gotit.pub
- Hugging Face
- IArxiv Recommender
- Machine learning based methods and apparatus for integrated circuit design delay calculation and verification
- Predicting Process-Level Behavior: A Benchmarking Study
- Predictive process monitoring
- ScienceCast
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →