Learning from the Storm: A Multivariate Machine Learning Approach to Predicting Hurricane-Induced Economic Losses

2025-06-22

AI Summary

Existing studies lack a unified framework integrating multidimensional factors for comprehensive assessment of hurricane-related economic losses. Method: This paper proposes the first three-dimensional machine learning framework—incorporating hurricane dynamic characteristics (e.g., intensity, track), hydrological environmental factors (e.g., storm surge, precipitation), and regional socioeconomic attributes (e.g., population density, building value)—to model and predict insured economic losses at the ZIP Code Tabulation Area (ZCTA) level. We employ zero-inflated Poisson regression coupled with explainable AI (XAI) techniques to quantify feature contributions and enhance both predictive accuracy and model interpretability. Contribution/Results: Experimental results demonstrate that our framework significantly outperforms conventional single-factor models and accurately identifies key risk drivers. The implementation code and curated dataset are publicly released, providing a reusable, transparent, and policy-relevant decision-support tool for coastal hazard risk assessment and resilience planning.

Technology Category

• Machine Learning: Multimodal Learning
• Humans and AI: Human-in-the-loop Machine Learning
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Machine learning and data science for the Web

Abstract

Florida is particularly vulnerable to hurricanes, which frequently cause substantial economic losses. While prior studies have explored specific contributors to hurricane-induced damage, few have developed a unified framework capable of integrating a broader range of influencing factors to comprehensively assess the sources of economic loss. In this study, we propose a comprehensive modeling framework that categorizes contributing factors into three key components: (1) hurricane characteristics, (2) water-related environmental factors, and (3) socioeconomic factors of affected areas. By integrating multi-source data and aggregating all variables at the finer spatial granularity of the ZIP Code Tabulation Area (ZCTA) level, we employ machine learning models to predict economic loss, using insurance claims as indicators of incurred damage. Beyond accurate loss prediction, our approach facilitates a systematic assessment of the relative importance of each component, providing practical guidance for disaster mitigation, risk assessment, and the development of adaptive urban strategies in coastal and storm-exposed areas. Our code is now available at: https://github.com/LabRAI/Hurricane-Induced-Economic-Loss-Prediction

Problem

• Predict hurricane-induced economic losses using machine learning
• Integrate hurricane, environmental, and socioeconomic factors
• Assess relative importance of factors for disaster mitigation

Innovation

• Multivariate machine learning for hurricane loss prediction
• ZCTA-level multi-source data integration
• Systematic assessment of contributing factors importance