×F2. Hybrid approaches to environmental data-driven modelling towards explainable intelligent systems

The session invites original contributions on a wide range of applications of statistical and analytical tools, artificial intelligence (AI), and machine learning techniques to environmental problems. A substantially “black box” nature of these techniques makes explanation and generalization of their results challenging. To address this problem, an explanation and visualization components can be integrated into analytical frameworks.

The techniques include, but are not limited to, exploratory [Big] data analysis, hybrid models and joint prediction and interpretation, intelligent data analysis and their combination with process-based simulations and computational modelling in such areas as environmental resource management, ecosystem services, land and forest use, agriculture, water cycle, air quality, and climate change.

The topics related to explainable AI techniques, or post-modelling explainability approaches for advanced analysis of environmental data are of special interest. Hybrid frameworks and techniques, success stories of their application and lessons learned are welcomed.

Key topics: Explainable modelling, Hybrid data analysis, Machine learning, Environmental sustainability