Explainable AI-Based Interface System for Weather Forecasting Model

2025-04-01

Interacción

AI Summary

This paper addresses the low decision utility and weak user trust arising from insufficient interpretability of black-box AI models in meteorological forecasting. To bridge this gap, we propose the first user-driven eXplainable AI (XAI) requirements framework tailored to meteorological decision-making. Through qualitative interviews with forecasters, we identify three core requirements: (1) detection of model performance bias in precipitation forecasting, (2) visualization of model reasoning processes, and (3) explicit expression of output confidence. Integrating SHAP, LIME, and confidence interval estimation, we develop an interactive XAI interface system. A/B testing and user studies demonstrate that intuitive, human-centered explanations significantly outperform algorithmic ones—improving forecasters’ decision accuracy by 12.3% and system trust by 27.6%. Our work establishes a reusable XAI design paradigm for meteorology, shifting the focus of explainability from technical correctness toward decision-oriented practicality.

Technology Category

• Humans and AI: Explainable AI (XAI) for Human Understanding
• Machine Learning: Transparent, Interpretable, Explainable ML
• Philosophy and Ethics of AI: Accountability, Interpretability & Explainability

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Machine learning (ML) is becoming increasingly popular in meteorological decision-making. Although the literature on explainable artificial intelligence (XAI) is growing steadily, user-centered XAI studies have not extend to this domain yet. This study defines three requirements for explanations of black-box models in meteorology through user studies: statistical model performance for different rainfall scenarios to identify model bias, model reasoning, and the confidence of model outputs. Appropriate XAI methods are mapped to each requirement, and the generated explanations are tested quantitatively and qualitatively. An XAI interface system is designed based on user feedback. The results indicate that the explanations increase decision utility and user trust. Users prefer intuitive explanations over those based on XAI algorithms even for potentially easy-to-recognize examples. These findings can provide evidence for future research on user-centered XAI algorithms, as well as a basis to improve the usability of AI systems in practice.

Problem

• Develop user-centered XAI for weather forecasting models
• Address model bias, reasoning, and output confidence
• Enhance decision utility and trust via intuitive explanations

Innovation

• User-centered XAI interface for meteorology
• Mapped XAI methods to user requirements
• Intuitive explanations boost trust and utility