Comparison of debris flow susceptibility assessment methods: support vector machine, particle swarm optimization, and feature selection techniques
The selection of important factors in machine learning-based susceptibility assessments is crucial to obtain reliable susceptibility results. In this study, metaheuristic optimization and feature selection techniques were applied to identify the most important input parameters for mapping debris flo...
Gespeichert in:
Veröffentlicht in: | Journal of mountain science 2024-02, Vol.21 (2), p.397-412 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 412 |
---|---|
container_issue | 2 |
container_start_page | 397 |
container_title | Journal of mountain science |
container_volume | 21 |
creator | Zhao, Haijun Wei, Aihua Ma, Fengshan Dai, Fenggang Jiang, Yongbing Li, Hui |
description | The selection of important factors in machine learning-based susceptibility assessments is crucial to obtain reliable susceptibility results. In this study, metaheuristic optimization and feature selection techniques were applied to identify the most important input parameters for mapping debris flow susceptibility in the southern mountain area of Chengde City in Hebei Province, China, by using machine learning algorithms. In total, 133 historical debris flow records and 16 related factors were selected. The support vector machine (SVM) was first used as the base classifier, and then a hybrid model was introduced by a two-step process. First, the particle swarm optimization (PSO) algorithm was employed to select the SVM model hyperparameters. Second, two feature selection algorithms, namely principal component analysis (PCA) and PSO, were integrated into the PSO-based SVM model, which generated the PCA-PSO-SVM and FS-PSO-SVM models, respectively. Three statistical metrics (accuracy, recall, and specificity) and the area under the receiver operating characteristic curve (AUC) were employed to evaluate and validate the performance of the models. The results indicated that the feature selection-based models exhibited the best performance, followed by the PSO-based SVM and SVM models. Moreover, the performance of the FS-PSO-SVM model was better than that of the PCA-PSO-SVM model, showing the highest AUC, accuracy, recall, and specificity values in both the training and testing processes. It was found that the selection of optimal features is crucial to improving the reliability of debris flow susceptibility assessment results. Moreover, the PSO algorithm was found to be not only an effective tool for hyperparameter optimization, but also a useful feature selection algorithm to improve prediction accuracies of debris flow susceptibility by using machine learning algorithms. The high and very high debris flow susceptibility zone appropriately covers 38.01% of the study area, where debris flow may occur under intensive human activities and heavy rainfall events. |
doi_str_mv | 10.1007/s11629-023-8395-9 |
format | Article |
fullrecord | <record><control><sourceid>wanfang_jour_proqu</sourceid><recordid>TN_cdi_wanfang_journals_sdkxxb_e202402003</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><wanfj_id>sdkxxb_e202402003</wanfj_id><sourcerecordid>sdkxxb_e202402003</sourcerecordid><originalsourceid>FETCH-LOGICAL-c302t-2ff05c9ebcbb84ffd253fcc04f9950d615c4b24c06f61d608c4f709971ef11113</originalsourceid><addsrcrecordid>eNp1kc-KFDEQxhtRcF19AG8BT8K2VpLu9MSbDP6DBS96Dul0ZSdrd9KmMs6uj-ETm6GFPZlLiqrf91XB1zQvObzhAMNb4lwJ3YKQ7U7qvtWPmguutWxBCv641moQrZJcPW2eEd0CqEHv-EXzZ5-W1eZAKbLk2YRjrZmf04nRkRyuJYxhDuWeWSIkWjAWtmA5pIneVWRdUy7sF7qSMlusO4SIV6w6luBmZHSyeWGpuizhty0hxStm48Q82nLMdY5zldY2K-gOMfw8Ij1vnng7E77491823z9--Lb_3F5__fRl__66dRJEaYX30DuNoxvHXef9JHrpnYPOa93DpHjvulF0DpRXfFKwc50fQOuBo-f1ycvm9eZ7stHbeGNu0zHHutHQ9OPubjQoQHQgAGRlX23smtP5xvIACy3qQqW5rhTfKJcTUUZv1hwWm-8NB3OOyWwxmRqTOcdkzhqxaaiy8Qbzg_P_RX8BNceZWw</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2929956919</pqid></control><display><type>article</type><title>Comparison of debris flow susceptibility assessment methods: support vector machine, particle swarm optimization, and feature selection techniques</title><source>SpringerNature Journals</source><source>Alma/SFX Local Collection</source><creator>Zhao, Haijun ; Wei, Aihua ; Ma, Fengshan ; Dai, Fenggang ; Jiang, Yongbing ; Li, Hui</creator><creatorcontrib>Zhao, Haijun ; Wei, Aihua ; Ma, Fengshan ; Dai, Fenggang ; Jiang, Yongbing ; Li, Hui</creatorcontrib><description>The selection of important factors in machine learning-based susceptibility assessments is crucial to obtain reliable susceptibility results. In this study, metaheuristic optimization and feature selection techniques were applied to identify the most important input parameters for mapping debris flow susceptibility in the southern mountain area of Chengde City in Hebei Province, China, by using machine learning algorithms. In total, 133 historical debris flow records and 16 related factors were selected. The support vector machine (SVM) was first used as the base classifier, and then a hybrid model was introduced by a two-step process. First, the particle swarm optimization (PSO) algorithm was employed to select the SVM model hyperparameters. Second, two feature selection algorithms, namely principal component analysis (PCA) and PSO, were integrated into the PSO-based SVM model, which generated the PCA-PSO-SVM and FS-PSO-SVM models, respectively. Three statistical metrics (accuracy, recall, and specificity) and the area under the receiver operating characteristic curve (AUC) were employed to evaluate and validate the performance of the models. The results indicated that the feature selection-based models exhibited the best performance, followed by the PSO-based SVM and SVM models. Moreover, the performance of the FS-PSO-SVM model was better than that of the PCA-PSO-SVM model, showing the highest AUC, accuracy, recall, and specificity values in both the training and testing processes. It was found that the selection of optimal features is crucial to improving the reliability of debris flow susceptibility assessment results. Moreover, the PSO algorithm was found to be not only an effective tool for hyperparameter optimization, but also a useful feature selection algorithm to improve prediction accuracies of debris flow susceptibility by using machine learning algorithms. The high and very high debris flow susceptibility zone appropriately covers 38.01% of the study area, where debris flow may occur under intensive human activities and heavy rainfall events.</description><identifier>ISSN: 1672-6316</identifier><identifier>EISSN: 1993-0321</identifier><identifier>EISSN: 1008-2786</identifier><identifier>DOI: 10.1007/s11629-023-8395-9</identifier><language>eng</language><publisher>Heidelberg: Science Press</publisher><subject>Accuracy ; Algorithms ; Debris flow ; Detritus ; Earth and Environmental Science ; Earth Sciences ; Ecology ; Environment ; Feature selection ; Flow mapping ; Geography ; Heuristic methods ; Learning algorithms ; Machine learning ; Mathematical models ; Mountain regions ; Mountainous areas ; Optimization ; Original Article ; Parameter identification ; Particle swarm optimization ; Principal components analysis ; Rainfall ; Recall ; Specificity ; Statistical analysis ; Support vector machines ; Susceptibility</subject><ispartof>Journal of mountain science, 2024-02, Vol.21 (2), p.397-412</ispartof><rights>Science Press, Institute of Mountain Hazards and Environment, CAS and Springer-Verlag GmbH Germany, part of Springer Nature 2024</rights><rights>Science Press, Institute of Mountain Hazards and Environment, CAS and Springer-Verlag GmbH Germany, part of Springer Nature 2024.</rights><rights>Copyright © Wanfang Data Co. Ltd. All Rights Reserved.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c302t-2ff05c9ebcbb84ffd253fcc04f9950d615c4b24c06f61d608c4f709971ef11113</cites><orcidid>0000-0001-5666-9860 ; 0000-0002-4585-6684 ; 0009-0004-0587-9947 ; 0009-0007-3536-0056 ; 0009-0005-0113-9551 ; 0000-0002-8698-7912</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Uhttp://www.wanfangdata.com.cn/images/PeriodicalImages/sdkxxb-e/sdkxxb-e.jpg</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/s11629-023-8395-9$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/s11629-023-8395-9$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>314,780,784,27924,27925,41488,42557,51319</link.rule.ids></links><search><creatorcontrib>Zhao, Haijun</creatorcontrib><creatorcontrib>Wei, Aihua</creatorcontrib><creatorcontrib>Ma, Fengshan</creatorcontrib><creatorcontrib>Dai, Fenggang</creatorcontrib><creatorcontrib>Jiang, Yongbing</creatorcontrib><creatorcontrib>Li, Hui</creatorcontrib><title>Comparison of debris flow susceptibility assessment methods: support vector machine, particle swarm optimization, and feature selection techniques</title><title>Journal of mountain science</title><addtitle>J. Mt. Sci</addtitle><description>The selection of important factors in machine learning-based susceptibility assessments is crucial to obtain reliable susceptibility results. In this study, metaheuristic optimization and feature selection techniques were applied to identify the most important input parameters for mapping debris flow susceptibility in the southern mountain area of Chengde City in Hebei Province, China, by using machine learning algorithms. In total, 133 historical debris flow records and 16 related factors were selected. The support vector machine (SVM) was first used as the base classifier, and then a hybrid model was introduced by a two-step process. First, the particle swarm optimization (PSO) algorithm was employed to select the SVM model hyperparameters. Second, two feature selection algorithms, namely principal component analysis (PCA) and PSO, were integrated into the PSO-based SVM model, which generated the PCA-PSO-SVM and FS-PSO-SVM models, respectively. Three statistical metrics (accuracy, recall, and specificity) and the area under the receiver operating characteristic curve (AUC) were employed to evaluate and validate the performance of the models. The results indicated that the feature selection-based models exhibited the best performance, followed by the PSO-based SVM and SVM models. Moreover, the performance of the FS-PSO-SVM model was better than that of the PCA-PSO-SVM model, showing the highest AUC, accuracy, recall, and specificity values in both the training and testing processes. It was found that the selection of optimal features is crucial to improving the reliability of debris flow susceptibility assessment results. Moreover, the PSO algorithm was found to be not only an effective tool for hyperparameter optimization, but also a useful feature selection algorithm to improve prediction accuracies of debris flow susceptibility by using machine learning algorithms. The high and very high debris flow susceptibility zone appropriately covers 38.01% of the study area, where debris flow may occur under intensive human activities and heavy rainfall events.</description><subject>Accuracy</subject><subject>Algorithms</subject><subject>Debris flow</subject><subject>Detritus</subject><subject>Earth and Environmental Science</subject><subject>Earth Sciences</subject><subject>Ecology</subject><subject>Environment</subject><subject>Feature selection</subject><subject>Flow mapping</subject><subject>Geography</subject><subject>Heuristic methods</subject><subject>Learning algorithms</subject><subject>Machine learning</subject><subject>Mathematical models</subject><subject>Mountain regions</subject><subject>Mountainous areas</subject><subject>Optimization</subject><subject>Original Article</subject><subject>Parameter identification</subject><subject>Particle swarm optimization</subject><subject>Principal components analysis</subject><subject>Rainfall</subject><subject>Recall</subject><subject>Specificity</subject><subject>Statistical analysis</subject><subject>Support vector machines</subject><subject>Susceptibility</subject><issn>1672-6316</issn><issn>1993-0321</issn><issn>1008-2786</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNp1kc-KFDEQxhtRcF19AG8BT8K2VpLu9MSbDP6DBS96Dul0ZSdrd9KmMs6uj-ETm6GFPZlLiqrf91XB1zQvObzhAMNb4lwJ3YKQ7U7qvtWPmguutWxBCv641moQrZJcPW2eEd0CqEHv-EXzZ5-W1eZAKbLk2YRjrZmf04nRkRyuJYxhDuWeWSIkWjAWtmA5pIneVWRdUy7sF7qSMlusO4SIV6w6luBmZHSyeWGpuizhty0hxStm48Q82nLMdY5zldY2K-gOMfw8Ij1vnng7E77491823z9--Lb_3F5__fRl__66dRJEaYX30DuNoxvHXef9JHrpnYPOa93DpHjvulF0DpRXfFKwc50fQOuBo-f1ycvm9eZ7stHbeGNu0zHHutHQ9OPubjQoQHQgAGRlX23smtP5xvIACy3qQqW5rhTfKJcTUUZv1hwWm-8NB3OOyWwxmRqTOcdkzhqxaaiy8Qbzg_P_RX8BNceZWw</recordid><startdate>20240201</startdate><enddate>20240201</enddate><creator>Zhao, Haijun</creator><creator>Wei, Aihua</creator><creator>Ma, Fengshan</creator><creator>Dai, Fenggang</creator><creator>Jiang, Yongbing</creator><creator>Li, Hui</creator><general>Science Press</general><general>Springer Nature B.V</general><general>School of Water Resources and Environment,Hebei GEO University,Shijiazhuang 050031,China%School of Water Resources and Environment,Hebei GEO University,Shijiazhuang 050031,China%The Sixth Geological Brigade of Geological Mineral Exploration and Development Bureau of Hebei Province,Shijiazhuang 050085,China</general><general>Innovation Academy for Earth Science,Chinese Academy of Sciences,Beijing 100029,China%Hebei Province Key Laboratory of Sustained Utilization&Development of Water Recourse,Hebei GEO University,Shijiazhuang 050031,China</general><general>Key Laboratory of Shale Gas and Geoengineering,Institute of Geology and Geophysics,Chinese Academy of Sciences,Beijing 100029,China</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7ST</scope><scope>7UA</scope><scope>C1K</scope><scope>F1W</scope><scope>H96</scope><scope>L.G</scope><scope>SOI</scope><scope>2B.</scope><scope>4A8</scope><scope>92I</scope><scope>93N</scope><scope>PSX</scope><scope>TCJ</scope><orcidid>https://orcid.org/0000-0001-5666-9860</orcidid><orcidid>https://orcid.org/0000-0002-4585-6684</orcidid><orcidid>https://orcid.org/0009-0004-0587-9947</orcidid><orcidid>https://orcid.org/0009-0007-3536-0056</orcidid><orcidid>https://orcid.org/0009-0005-0113-9551</orcidid><orcidid>https://orcid.org/0000-0002-8698-7912</orcidid></search><sort><creationdate>20240201</creationdate><title>Comparison of debris flow susceptibility assessment methods: support vector machine, particle swarm optimization, and feature selection techniques</title><author>Zhao, Haijun ; Wei, Aihua ; Ma, Fengshan ; Dai, Fenggang ; Jiang, Yongbing ; Li, Hui</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c302t-2ff05c9ebcbb84ffd253fcc04f9950d615c4b24c06f61d608c4f709971ef11113</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Accuracy</topic><topic>Algorithms</topic><topic>Debris flow</topic><topic>Detritus</topic><topic>Earth and Environmental Science</topic><topic>Earth Sciences</topic><topic>Ecology</topic><topic>Environment</topic><topic>Feature selection</topic><topic>Flow mapping</topic><topic>Geography</topic><topic>Heuristic methods</topic><topic>Learning algorithms</topic><topic>Machine learning</topic><topic>Mathematical models</topic><topic>Mountain regions</topic><topic>Mountainous areas</topic><topic>Optimization</topic><topic>Original Article</topic><topic>Parameter identification</topic><topic>Particle swarm optimization</topic><topic>Principal components analysis</topic><topic>Rainfall</topic><topic>Recall</topic><topic>Specificity</topic><topic>Statistical analysis</topic><topic>Support vector machines</topic><topic>Susceptibility</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Zhao, Haijun</creatorcontrib><creatorcontrib>Wei, Aihua</creatorcontrib><creatorcontrib>Ma, Fengshan</creatorcontrib><creatorcontrib>Dai, Fenggang</creatorcontrib><creatorcontrib>Jiang, Yongbing</creatorcontrib><creatorcontrib>Li, Hui</creatorcontrib><collection>CrossRef</collection><collection>Environment Abstracts</collection><collection>Water Resources Abstracts</collection><collection>Environmental Sciences and Pollution Management</collection><collection>ASFA: Aquatic Sciences and Fisheries Abstracts</collection><collection>Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources</collection><collection>Aquatic Science & Fisheries Abstracts (ASFA) Professional</collection><collection>Environment Abstracts</collection><collection>Wanfang Data Journals - Hong Kong</collection><collection>WANFANG Data Centre</collection><collection>Wanfang Data Journals</collection><collection>万方数据期刊 - 香港版</collection><collection>China Online Journals (COJ)</collection><collection>China Online Journals (COJ)</collection><jtitle>Journal of mountain science</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Zhao, Haijun</au><au>Wei, Aihua</au><au>Ma, Fengshan</au><au>Dai, Fenggang</au><au>Jiang, Yongbing</au><au>Li, Hui</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Comparison of debris flow susceptibility assessment methods: support vector machine, particle swarm optimization, and feature selection techniques</atitle><jtitle>Journal of mountain science</jtitle><stitle>J. Mt. Sci</stitle><date>2024-02-01</date><risdate>2024</risdate><volume>21</volume><issue>2</issue><spage>397</spage><epage>412</epage><pages>397-412</pages><issn>1672-6316</issn><eissn>1993-0321</eissn><eissn>1008-2786</eissn><abstract>The selection of important factors in machine learning-based susceptibility assessments is crucial to obtain reliable susceptibility results. In this study, metaheuristic optimization and feature selection techniques were applied to identify the most important input parameters for mapping debris flow susceptibility in the southern mountain area of Chengde City in Hebei Province, China, by using machine learning algorithms. In total, 133 historical debris flow records and 16 related factors were selected. The support vector machine (SVM) was first used as the base classifier, and then a hybrid model was introduced by a two-step process. First, the particle swarm optimization (PSO) algorithm was employed to select the SVM model hyperparameters. Second, two feature selection algorithms, namely principal component analysis (PCA) and PSO, were integrated into the PSO-based SVM model, which generated the PCA-PSO-SVM and FS-PSO-SVM models, respectively. Three statistical metrics (accuracy, recall, and specificity) and the area under the receiver operating characteristic curve (AUC) were employed to evaluate and validate the performance of the models. The results indicated that the feature selection-based models exhibited the best performance, followed by the PSO-based SVM and SVM models. Moreover, the performance of the FS-PSO-SVM model was better than that of the PCA-PSO-SVM model, showing the highest AUC, accuracy, recall, and specificity values in both the training and testing processes. It was found that the selection of optimal features is crucial to improving the reliability of debris flow susceptibility assessment results. Moreover, the PSO algorithm was found to be not only an effective tool for hyperparameter optimization, but also a useful feature selection algorithm to improve prediction accuracies of debris flow susceptibility by using machine learning algorithms. The high and very high debris flow susceptibility zone appropriately covers 38.01% of the study area, where debris flow may occur under intensive human activities and heavy rainfall events.</abstract><cop>Heidelberg</cop><pub>Science Press</pub><doi>10.1007/s11629-023-8395-9</doi><tpages>16</tpages><orcidid>https://orcid.org/0000-0001-5666-9860</orcidid><orcidid>https://orcid.org/0000-0002-4585-6684</orcidid><orcidid>https://orcid.org/0009-0004-0587-9947</orcidid><orcidid>https://orcid.org/0009-0007-3536-0056</orcidid><orcidid>https://orcid.org/0009-0005-0113-9551</orcidid><orcidid>https://orcid.org/0000-0002-8698-7912</orcidid></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1672-6316 |
ispartof | Journal of mountain science, 2024-02, Vol.21 (2), p.397-412 |
issn | 1672-6316 1993-0321 1008-2786 |
language | eng |
recordid | cdi_wanfang_journals_sdkxxb_e202402003 |
source | SpringerNature Journals; Alma/SFX Local Collection |
subjects | Accuracy Algorithms Debris flow Detritus Earth and Environmental Science Earth Sciences Ecology Environment Feature selection Flow mapping Geography Heuristic methods Learning algorithms Machine learning Mathematical models Mountain regions Mountainous areas Optimization Original Article Parameter identification Particle swarm optimization Principal components analysis Rainfall Recall Specificity Statistical analysis Support vector machines Susceptibility |
title | Comparison of debris flow susceptibility assessment methods: support vector machine, particle swarm optimization, and feature selection techniques |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T19%3A19%3A30IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-wanfang_jour_proqu&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Comparison%20of%20debris%20flow%20susceptibility%20assessment%20methods:%20support%20vector%20machine,%20particle%20swarm%20optimization,%20and%20feature%20selection%20techniques&rft.jtitle=Journal%20of%20mountain%20science&rft.au=Zhao,%20Haijun&rft.date=2024-02-01&rft.volume=21&rft.issue=2&rft.spage=397&rft.epage=412&rft.pages=397-412&rft.issn=1672-6316&rft.eissn=1993-0321&rft_id=info:doi/10.1007/s11629-023-8395-9&rft_dat=%3Cwanfang_jour_proqu%3Esdkxxb_e202402003%3C/wanfang_jour_proqu%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2929956919&rft_id=info:pmid/&rft_wanfj_id=sdkxxb_e202402003&rfr_iscdi=true |