Explainable Artificial Intelligence (XAI) for Flood Susceptibility Assessment in Seoul: Leveraging Evolutionary and Bayesian AutoML Optimization
This study aims to enhance the accuracy and interpretability of flood susceptibility mapping (FSM) in Seoul, South Korea, by integrating automated machine learning (AutoML) with explainable artificial intelligence (XAI) techniques. Ten topographic and environmental conditioning factors were selected...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/13/2244 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850115973358551040 |
|---|---|
| author | Kounghoon Nam Youngkyu Lee Sungsu Lee Sungyoon Kim Shuai Zhang |
| author_facet | Kounghoon Nam Youngkyu Lee Sungsu Lee Sungyoon Kim Shuai Zhang |
| author_sort | Kounghoon Nam |
| collection | DOAJ |
| description | This study aims to enhance the accuracy and interpretability of flood susceptibility mapping (FSM) in Seoul, South Korea, by integrating automated machine learning (AutoML) with explainable artificial intelligence (XAI) techniques. Ten topographic and environmental conditioning factors were selected as model inputs. We first employed the Tree-based Pipeline Optimization Tool (TPOT), an evolutionary AutoML algorithm, to construct baseline ensemble models using Gradient Boosting (GB), Random Forest (RF), and XGBoost (XGB). These models were further fine-tuned using Bayesian optimization via Optuna. To interpret the model outcomes, SHAP (SHapley Additive exPlanations) was applied to analyze both the global and local contributions of each factor. The SHAP analysis revealed that lower elevation, slope, and stream distance, as well as higher stream density and built-up areas, were the most influential factors contributing to flood susceptibility. Moreover, interactions between these factors, such as built-up areas located on gentle slopes near streams, further intensified flood risk. The susceptibility maps were reclassified into five categories (very low to very high), and the GB model identified that approximately 15.047% of the study area falls under very-high-flood-risk zones. Among the models, the GB classifier achieved the highest performance, followed by XGB and RF. The proposed framework, which integrates TPOT, Optuna, and SHAP within an XAI pipeline, not only improves predictive capability but also offers transparent insights into feature behavior and model logic. These findings support more robust and interpretable flood risk assessments for effective disaster management in urban areas. |
| format | Article |
| id | doaj-art-1fd88f065dd44d3f99ad22a58c291ecf |
| institution | OA Journals |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-1fd88f065dd44d3f99ad22a58c291ecf2025-08-20T02:36:27ZengMDPI AGRemote Sensing2072-42922025-06-011713224410.3390/rs17132244Explainable Artificial Intelligence (XAI) for Flood Susceptibility Assessment in Seoul: Leveraging Evolutionary and Bayesian AutoML OptimizationKounghoon Nam0Youngkyu Lee1Sungsu Lee2Sungyoon Kim3Shuai Zhang4Fire Insurers Laboratories of Korea, Yeoju 12661, Republic of KoreaFire Insurers Laboratories of Korea, Yeoju 12661, Republic of KoreaDepartment of Civil Engineering, Chungbuk National University, Cheongju 28644, Republic of KoreaFire Insurers Laboratories of Korea, Yeoju 12661, Republic of KoreaInstitute of Geomechanics, Chinese Academy of Geological Sciences, Beijing 100081, ChinaThis study aims to enhance the accuracy and interpretability of flood susceptibility mapping (FSM) in Seoul, South Korea, by integrating automated machine learning (AutoML) with explainable artificial intelligence (XAI) techniques. Ten topographic and environmental conditioning factors were selected as model inputs. We first employed the Tree-based Pipeline Optimization Tool (TPOT), an evolutionary AutoML algorithm, to construct baseline ensemble models using Gradient Boosting (GB), Random Forest (RF), and XGBoost (XGB). These models were further fine-tuned using Bayesian optimization via Optuna. To interpret the model outcomes, SHAP (SHapley Additive exPlanations) was applied to analyze both the global and local contributions of each factor. The SHAP analysis revealed that lower elevation, slope, and stream distance, as well as higher stream density and built-up areas, were the most influential factors contributing to flood susceptibility. Moreover, interactions between these factors, such as built-up areas located on gentle slopes near streams, further intensified flood risk. The susceptibility maps were reclassified into five categories (very low to very high), and the GB model identified that approximately 15.047% of the study area falls under very-high-flood-risk zones. Among the models, the GB classifier achieved the highest performance, followed by XGB and RF. The proposed framework, which integrates TPOT, Optuna, and SHAP within an XAI pipeline, not only improves predictive capability but also offers transparent insights into feature behavior and model logic. These findings support more robust and interpretable flood risk assessments for effective disaster management in urban areas.https://www.mdpi.com/2072-4292/17/13/2244explainable artificial intelligence (XAI)TPOT (evolutionary optimization)Optuna (bayesian optimization)SHAP interpretationhyperparameter tuning |
| spellingShingle | Kounghoon Nam Youngkyu Lee Sungsu Lee Sungyoon Kim Shuai Zhang Explainable Artificial Intelligence (XAI) for Flood Susceptibility Assessment in Seoul: Leveraging Evolutionary and Bayesian AutoML Optimization Remote Sensing explainable artificial intelligence (XAI) TPOT (evolutionary optimization) Optuna (bayesian optimization) SHAP interpretation hyperparameter tuning |
| title | Explainable Artificial Intelligence (XAI) for Flood Susceptibility Assessment in Seoul: Leveraging Evolutionary and Bayesian AutoML Optimization |
| title_full | Explainable Artificial Intelligence (XAI) for Flood Susceptibility Assessment in Seoul: Leveraging Evolutionary and Bayesian AutoML Optimization |
| title_fullStr | Explainable Artificial Intelligence (XAI) for Flood Susceptibility Assessment in Seoul: Leveraging Evolutionary and Bayesian AutoML Optimization |
| title_full_unstemmed | Explainable Artificial Intelligence (XAI) for Flood Susceptibility Assessment in Seoul: Leveraging Evolutionary and Bayesian AutoML Optimization |
| title_short | Explainable Artificial Intelligence (XAI) for Flood Susceptibility Assessment in Seoul: Leveraging Evolutionary and Bayesian AutoML Optimization |
| title_sort | explainable artificial intelligence xai for flood susceptibility assessment in seoul leveraging evolutionary and bayesian automl optimization |
| topic | explainable artificial intelligence (XAI) TPOT (evolutionary optimization) Optuna (bayesian optimization) SHAP interpretation hyperparameter tuning |
| url | https://www.mdpi.com/2072-4292/17/13/2244 |
| work_keys_str_mv | AT kounghoonnam explainableartificialintelligencexaiforfloodsusceptibilityassessmentinseoulleveragingevolutionaryandbayesianautomloptimization AT youngkyulee explainableartificialintelligencexaiforfloodsusceptibilityassessmentinseoulleveragingevolutionaryandbayesianautomloptimization AT sungsulee explainableartificialintelligencexaiforfloodsusceptibilityassessmentinseoulleveragingevolutionaryandbayesianautomloptimization AT sungyoonkim explainableartificialintelligencexaiforfloodsusceptibilityassessmentinseoulleveragingevolutionaryandbayesianautomloptimization AT shuaizhang explainableartificialintelligencexaiforfloodsusceptibilityassessmentinseoulleveragingevolutionaryandbayesianautomloptimization |