Flood-prone area mapping using a synergistic approach with swarm intelligence and gradient boosting algorithms

Accurate flood susceptibility mapping (FSM) is a control approach to flood management. This research introduces a novel approach to increase the accuracy of FSM by optimizing the CatBoost algorithm with two swarm-based metaheuristic algorithms: the Zebra optimization algorithm (ZOA) and the Whale op...

Full description

Saved in:
Bibliographic Details
Published in:Scientific reports Vol. 15; no. 1; pp. 27924 - 20
Main Authors: Razavi-Termeh, Seyed Vahid, Sadeghi-Niaraki, Abolghasem, Abba, Sani I., Hussain, Jamil, Choi, Soo-Mi
Format: Journal Article
Language:English
Published: London Nature Publishing Group UK 31.07.2025
Nature Publishing Group
Nature Portfolio
Subjects:
704/242
704/4111
Accuracy
Algorithms
Artificial intelligence
Boosting algorithm
Flood control
Flood management
Flood-prone area mapping
Floods
Geographic information systems
Humanities and Social Sciences
Infrastructure
Learning algorithms
Machine learning
Mapping
multidisciplinary
Neural networks
Optimization algorithms
Science
Science (multidisciplinary)
Semi-arid climate
Storm damage
Support vector machines
Swarm-based metaheuristics
Topography
Wind
Middle East
Iran
Semi-arid climate
Flood management
Flood-prone area mapping
Boosting algorithm
Swarm-based metaheuristics
ISSN:2045-2322, 2045-2322
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Accurate flood susceptibility mapping (FSM) is a control approach to flood management. This research introduces a novel approach to increase the accuracy of FSM by optimizing the CatBoost algorithm with two swarm-based metaheuristic algorithms: the Zebra optimization algorithm (ZOA) and the Whale optimization algorithm (WOA). This research addresses the critical gap in determining optimal hyperparameters for machine learning models in FSM. Existing studies often ignore the importance of hyperparameter tuning in FSM, leading to suboptimal model performance. This research seeks to enhance and improve the accuracy of FSM in Shushtar County, located in southwest Iran, by utilizing the CatBoost-WOA and CatBoost-ZOA algorithms. In this research, FSM was conducted by using 13 parameters that affect floods and flood occurrence points as inputs. The evaluation results of the flood susceptibility maps showed an accuracy of 84.2% for CatBoost, 85% for CatBoost-WOA, and 87.2% for CatBoost-ZOA. This represents a 3.0% absolute improvement in accuracy with the ZOA-optimized model over the non-optimized CatBoost. The results of this study demonstrated that integrating swarm-based methods with machine learning boosting algorithms significantly enhanced FSM accuracy. The results of this study, as a non-structural approach, can help managers and decision-makers in flood management and control.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-025-12022-6