Predictive Monitoring

Predictive monitoring aims to anticipate future system states or events based on current observations, enabling proactive interventions and improved decision-making. Current research heavily utilizes machine learning, focusing on recurrent neural networks (like LSTMs), reinforcement learning, and graph convolutional networks, often incorporating techniques like conformal prediction for uncertainty quantification and Shapley values for explainability. This field is significant for its applications across diverse domains, from healthcare and industrial process optimization to environmental monitoring, offering the potential for increased efficiency, safety, and resource management.

Papers