Reproduce “Interpretable and Explainable Machine Learning Methods for Predictive Process Monitoring: A Systematic Literature Review” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Nijat Mehdiyev, Maxim Majlatow, Peter Fettke
A systematic literature review comprehensively maps existing research on interpretable and explainable machine learning methods in predictive process monitoring, categorizing methodologies, application contexts, and evaluation practices. The work identifies prevalent trends and critical gaps in integrating explainable AI within process analytics.
Key insights
- Research in XAI for PPM is a rapidly growing field, evidenced by a significant increase in publications, particularly since 2020.
- A clear distinction exists between intrinsically interpretable (white-box) models, often simpler, and complex black-box models that rely on post-hoc explanation techniques.
- Most studies lack formal evaluation of their XAI methods, frequently prioritizing model accuracy over thorough assessment of explanation quality or user understanding, indicating a critical gap in research practices.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.