LLM-driven semantic explanations for soil moisture prediction models

Efficient soil moisture prediction is crucial for sustainable agricultural practices, especially in the face of climate change and increasing water scarcity. However, the adoption of machine learning (ML) models in this context is frequently limited by their lack of interpretability, particularly among non-expert users such as farmers. This study proposes a novel approach to soil moisture prediction that combines high predictive performance with enhanced explainability. We propose a framework that leverages large language models (LLMs) to generate textual explanations based on a proposed irrigation and soil moisture ontology, thus making the model's predictions more understandable to farmers. The ontology formalizes essential agricultural concepts and their interrelationships, enabling semantically rich explanations to bridge the gap between sophisticated model results and practical decision-making. Our approach is exemplified by a prototype system that provides both predictions and user-friendly explanations. The findings highlight the potential of combining advanced ML techniques with semantic reasoning to improve the interpretability and adoption of Artificial Intelligence in agriculture.