Novel Ensembles of Deep Learning Neural Network and Statistical Learning for Flash-Flood Susceptibility Mapping
This study aimed to assess flash-flood susceptibility using a new hybridization approach of Deep Neural Network (DNN), Analytical Hierarchy Process (AHP), and Frequency Ratio (FR). A catchment area in south-eastern Romania was selected for this proposed approach. In this regard, a geospatial databas...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-05-01
|
Series: | Water |
Subjects: | |
Online Access: | https://www.mdpi.com/2073-4441/12/6/1549 |
_version_ | 1797566708731346944 |
---|---|
author | Romulus Costache Phuong Thao Thi Ngo Dieu Tien Bui |
author_facet | Romulus Costache Phuong Thao Thi Ngo Dieu Tien Bui |
author_sort | Romulus Costache |
collection | DOAJ |
description | This study aimed to assess flash-flood susceptibility using a new hybridization approach of Deep Neural Network (DNN), Analytical Hierarchy Process (AHP), and Frequency Ratio (FR). A catchment area in south-eastern Romania was selected for this proposed approach. In this regard, a geospatial database of the flood with 178 flood locations and with 10 flash-flood predictors was prepared and used for this proposed approach. AHP and FR were used for processing and coding the predictors into a numeric format, whereas DNN, which is a powerful and state-of-the-art probabilistic machine leaning, was employed to build an inference flash-flood model. The reliability of the models was verified with the help of Receiver Operating Characteristic (ROC) Curve, Area Under Curve (AUC), and several statistical measures. The result shows that the two proposed ensemble models, DNN-AHP and DNN-FR, are capable of predicting future flash-flood areas with accuracy higher than 92%; therefore, they are a new tool for flash-flood studies. |
first_indexed | 2024-03-10T19:31:17Z |
format | Article |
id | doaj.art-125a8b0b9af24a0f950af45e5a3be3b8 |
institution | Directory Open Access Journal |
issn | 2073-4441 |
language | English |
last_indexed | 2024-03-10T19:31:17Z |
publishDate | 2020-05-01 |
publisher | MDPI AG |
record_format | Article |
series | Water |
spelling | doaj.art-125a8b0b9af24a0f950af45e5a3be3b82023-11-20T02:06:26ZengMDPI AGWater2073-44412020-05-01126154910.3390/w12061549Novel Ensembles of Deep Learning Neural Network and Statistical Learning for Flash-Flood Susceptibility MappingRomulus Costache0Phuong Thao Thi Ngo1Dieu Tien Bui2Research Institute of the University of Bucharest, 90-92 Sos. Panduri, 5th District, 050663 Bucharest, RomaniaInstitute of Research and Development, Duy Tan University, Da Nang 550000, VietnamGeographic Information System Group, Department of Business and IT, University of South-Eastern Norway, N-3800 Bø i Telemark, NorwayThis study aimed to assess flash-flood susceptibility using a new hybridization approach of Deep Neural Network (DNN), Analytical Hierarchy Process (AHP), and Frequency Ratio (FR). A catchment area in south-eastern Romania was selected for this proposed approach. In this regard, a geospatial database of the flood with 178 flood locations and with 10 flash-flood predictors was prepared and used for this proposed approach. AHP and FR were used for processing and coding the predictors into a numeric format, whereas DNN, which is a powerful and state-of-the-art probabilistic machine leaning, was employed to build an inference flash-flood model. The reliability of the models was verified with the help of Receiver Operating Characteristic (ROC) Curve, Area Under Curve (AUC), and several statistical measures. The result shows that the two proposed ensemble models, DNN-AHP and DNN-FR, are capable of predicting future flash-flood areas with accuracy higher than 92%; therefore, they are a new tool for flash-flood studies.https://www.mdpi.com/2073-4441/12/6/1549flash floodspatial modelingdeep learningstatistical learningRomania |
spellingShingle | Romulus Costache Phuong Thao Thi Ngo Dieu Tien Bui Novel Ensembles of Deep Learning Neural Network and Statistical Learning for Flash-Flood Susceptibility Mapping Water flash flood spatial modeling deep learning statistical learning Romania |
title | Novel Ensembles of Deep Learning Neural Network and Statistical Learning for Flash-Flood Susceptibility Mapping |
title_full | Novel Ensembles of Deep Learning Neural Network and Statistical Learning for Flash-Flood Susceptibility Mapping |
title_fullStr | Novel Ensembles of Deep Learning Neural Network and Statistical Learning for Flash-Flood Susceptibility Mapping |
title_full_unstemmed | Novel Ensembles of Deep Learning Neural Network and Statistical Learning for Flash-Flood Susceptibility Mapping |
title_short | Novel Ensembles of Deep Learning Neural Network and Statistical Learning for Flash-Flood Susceptibility Mapping |
title_sort | novel ensembles of deep learning neural network and statistical learning for flash flood susceptibility mapping |
topic | flash flood spatial modeling deep learning statistical learning Romania |
url | https://www.mdpi.com/2073-4441/12/6/1549 |
work_keys_str_mv | AT romuluscostache novelensemblesofdeeplearningneuralnetworkandstatisticallearningforflashfloodsusceptibilitymapping AT phuongthaothingo novelensemblesofdeeplearningneuralnetworkandstatisticallearningforflashfloodsusceptibilitymapping AT dieutienbui novelensemblesofdeeplearningneuralnetworkandstatisticallearningforflashfloodsusceptibilitymapping |