Machine learning prediction of climate-induced disaster injuries

May Haggag (), Eman Rezk () and Wael El-Dakhakhni ()
Additional contact information
May Haggag: McMaster Institute for Multi-Hazard Systemic Risk Studies (INTERFACE)
Eman Rezk: McMaster University
Wael El-Dakhakhni: McMaster University

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards, 2023, vol. 116, issue 3, No 36, 3645-3667

Abstract: Abstract The frequency of climate-induced disasters (CID) has exhibited a fivefold increase in the last five decades. In terms of CID global impact, around 1.7 billion people were affected in the last decade, and in 2020 alone, 30 million people were displaced due to CID. Furthermore, over the past two decades, 1 million deaths were reported and over $1.7 trillion in damage was attributed to CID. As such, the World Economic Forum, in its 2022 report, has identified climate action failure and extreme weather as the two most severe global risks to be considered over the next decade. Given the uncertainty and complexity associated with predicting CID frequencies and related impacts, the use of descriptive-, predictive- and prescriptive data analytics is key. To demonstrate the power of data analytics in predicting CID impacts, this work focuses on developing a data-driven machine learning model that predicts tornado-induced injuries based on a diverse set of input features ranging from hazard-, social-, geographic-, and climate-related features together with attributes related to community vulnerability, risk and resilience. These input features are then used to train and test various machine learning-based prediction models utilizing diverse techniques including decision trees, ensemble methods, and artificial neural networks. These models are subsequently evaluated to select the most significant features and the best performing model. In addition, several variable importance techniques are used to evaluate the dominance of all features and develop a model considering the most influential features. The results show that the best performing model had a testing accuracy of 83%. In addition, the results highlighted the apparent relationship between hazard-related attributes and tornadoes’ injury predictions. The developed approach is a step forward in harnessing the power of machine learning for improving our adaptation, preparedness, and planning towards global CID resilience.

Keywords: Climate-induced disasters; Data analytics; Injuries; Machine learning; Resilience (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11069-023-05829-x Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:nathaz:v:116:y:2023:i:3:d:10.1007_s11069-023-05829-x

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11069

DOI: 10.1007/s11069-023-05829-x

Access Statistics for this article

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards is currently edited by Thomas Glade, Tad S. Murty and Vladimír Schenk

More articles in Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards from Springer, International Society for the Prevention and Mitigation of Natural Hazards
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().