Intelligent regional subsurface prediction based on limited borehole data and interpretability stacking technique of ensemble learning

This study introduces an intelligent method for regional subsurface prediction using a Stacking ensemble learning approach, which incorporates K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), Gradient Boosted Decision Trees (GBDT), and Xgboost as base classifiers, with Logistic Reg...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Bulletin of engineering geology and the environment 2024-07, Vol.83 (7), p.272, Article 272
Hauptverfasser: Bai, Jun, Wang, Sheng, Xu, Qiang, Zhu, Junsheng, Li, Zhaoqi, Lai, Kun, Liu, Xingyi, Chen, Zongjie
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue 7
container_start_page 272
container_title Bulletin of engineering geology and the environment
container_volume 83
creator Bai, Jun
Wang, Sheng
Xu, Qiang
Zhu, Junsheng
Li, Zhaoqi
Lai, Kun
Liu, Xingyi
Chen, Zongjie
description This study introduces an intelligent method for regional subsurface prediction using a Stacking ensemble learning approach, which incorporates K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), Gradient Boosted Decision Trees (GBDT), and Xgboost as base classifiers, with Logistic Regression (LR) serving as the meta-classifier. Leveraging data from 1119 boreholes in Zigong City, China, this method achieves a prediction accuracy of 93%, and notably improves the prediction of weak layers, with accuracy rates ranging from 71.4% to 81.5%. This enhancement is particularly significant in areas with a random distribution of excavation and backfill. Furthermore, this study employs the SHAP method (SHapley Additive explanations) to interpret the Stacking ensemble learning model, revealing that the outputs of the base classifiers enhance the feature set for the meta-classifier, effectively addressing the insensitivity of the spatial coordinates x, y, and z as input features for lithology prediction. The findings demonstrate that the expansion of effective feature dimensions is key to the superior performance of the Stacking ensemble learning method in regional subsurface lithology prediction.
doi_str_mv 10.1007/s10064-024-03758-y
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_3066864937</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3066598283</sourcerecordid><originalsourceid>FETCH-LOGICAL-c228t-519dad108be1a9d7ef265bbdec46f7a632fbb001ac9b93ed3fc9812cc7af5cd53</originalsourceid><addsrcrecordid>eNqFkctOAyEYhSdGE2v1BVyRuB7lUmaGpWm8NGniRteEy09LnTIV6GJewOcWrdGdLoATON8Jf05VXRJ8TTBub1LZm1mNaVms5V09HlUTMmO8Fpy1xz-aitPqLKUNxoR3lEyq90XI0Pd-BSGjCCs_BNWjtNdpH50ygHYRrDe53COtElhURO-3PhephwjroQdkVVZIBYt8SYsFyUr73ucRpazMqw8rlMGsg3_bAxocgpBgqwvYg4qhPJ9XJ071CS6-z2n1cn_3PH-sl08Pi_ntsjaUdrnmRFhlCe40ECVsC442XGsLZta4VjWMOq3LbMoILRhY5ozoCDWmVY4by9m0ujrk7uJQ_pKy3Az7WEZOkuGm6ZqZYO1_Li462rHiogeXiUNKEZzcRb9VcZQEy89W5KEVWVqRX63IsUDsAKViDiuIv9F_UB9wxJR_</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3066598283</pqid></control><display><type>article</type><title>Intelligent regional subsurface prediction based on limited borehole data and interpretability stacking technique of ensemble learning</title><source>SpringerLink (Online service)</source><creator>Bai, Jun ; Wang, Sheng ; Xu, Qiang ; Zhu, Junsheng ; Li, Zhaoqi ; Lai, Kun ; Liu, Xingyi ; Chen, Zongjie</creator><creatorcontrib>Bai, Jun ; Wang, Sheng ; Xu, Qiang ; Zhu, Junsheng ; Li, Zhaoqi ; Lai, Kun ; Liu, Xingyi ; Chen, Zongjie</creatorcontrib><description>This study introduces an intelligent method for regional subsurface prediction using a Stacking ensemble learning approach, which incorporates K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), Gradient Boosted Decision Trees (GBDT), and Xgboost as base classifiers, with Logistic Regression (LR) serving as the meta-classifier. Leveraging data from 1119 boreholes in Zigong City, China, this method achieves a prediction accuracy of 93%, and notably improves the prediction of weak layers, with accuracy rates ranging from 71.4% to 81.5%. This enhancement is particularly significant in areas with a random distribution of excavation and backfill. Furthermore, this study employs the SHAP method (SHapley Additive explanations) to interpret the Stacking ensemble learning model, revealing that the outputs of the base classifiers enhance the feature set for the meta-classifier, effectively addressing the insensitivity of the spatial coordinates x, y, and z as input features for lithology prediction. The findings demonstrate that the expansion of effective feature dimensions is key to the superior performance of the Stacking ensemble learning method in regional subsurface lithology prediction.</description><identifier>ISSN: 1435-9529</identifier><identifier>EISSN: 1435-9537</identifier><identifier>DOI: 10.1007/s10064-024-03758-y</identifier><language>eng</language><publisher>Berlin/Heidelberg: Springer Berlin Heidelberg</publisher><subject>Accuracy ; Algorithms ; Artificial intelligence ; Boreholes ; Classifiers ; Datasets ; Decision trees ; Dredging ; Earth and Environmental Science ; Earth Sciences ; Ensemble learning ; Excavation ; Foundations ; Geoecology/Natural Processes ; Geoengineering ; Geology ; Geotechnical Engineering &amp; Applied Earth Sciences ; Hydraulics ; Learning ; Lithology ; Machine learning ; Nature Conservation ; Original Paper ; Predictions ; Regression analysis</subject><ispartof>Bulletin of engineering geology and the environment, 2024-07, Vol.83 (7), p.272, Article 272</ispartof><rights>The Author(s) 2024</rights><rights>The Author(s) 2024.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c228t-519dad108be1a9d7ef265bbdec46f7a632fbb001ac9b93ed3fc9812cc7af5cd53</cites><orcidid>0000-0003-1479-0309</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/s10064-024-03758-y$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/s10064-024-03758-y$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>314,780,784,27924,27925,41488,42557,51319</link.rule.ids></links><search><creatorcontrib>Bai, Jun</creatorcontrib><creatorcontrib>Wang, Sheng</creatorcontrib><creatorcontrib>Xu, Qiang</creatorcontrib><creatorcontrib>Zhu, Junsheng</creatorcontrib><creatorcontrib>Li, Zhaoqi</creatorcontrib><creatorcontrib>Lai, Kun</creatorcontrib><creatorcontrib>Liu, Xingyi</creatorcontrib><creatorcontrib>Chen, Zongjie</creatorcontrib><title>Intelligent regional subsurface prediction based on limited borehole data and interpretability stacking technique of ensemble learning</title><title>Bulletin of engineering geology and the environment</title><addtitle>Bull Eng Geol Environ</addtitle><description>This study introduces an intelligent method for regional subsurface prediction using a Stacking ensemble learning approach, which incorporates K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), Gradient Boosted Decision Trees (GBDT), and Xgboost as base classifiers, with Logistic Regression (LR) serving as the meta-classifier. Leveraging data from 1119 boreholes in Zigong City, China, this method achieves a prediction accuracy of 93%, and notably improves the prediction of weak layers, with accuracy rates ranging from 71.4% to 81.5%. This enhancement is particularly significant in areas with a random distribution of excavation and backfill. Furthermore, this study employs the SHAP method (SHapley Additive explanations) to interpret the Stacking ensemble learning model, revealing that the outputs of the base classifiers enhance the feature set for the meta-classifier, effectively addressing the insensitivity of the spatial coordinates x, y, and z as input features for lithology prediction. The findings demonstrate that the expansion of effective feature dimensions is key to the superior performance of the Stacking ensemble learning method in regional subsurface lithology prediction.</description><subject>Accuracy</subject><subject>Algorithms</subject><subject>Artificial intelligence</subject><subject>Boreholes</subject><subject>Classifiers</subject><subject>Datasets</subject><subject>Decision trees</subject><subject>Dredging</subject><subject>Earth and Environmental Science</subject><subject>Earth Sciences</subject><subject>Ensemble learning</subject><subject>Excavation</subject><subject>Foundations</subject><subject>Geoecology/Natural Processes</subject><subject>Geoengineering</subject><subject>Geology</subject><subject>Geotechnical Engineering &amp; Applied Earth Sciences</subject><subject>Hydraulics</subject><subject>Learning</subject><subject>Lithology</subject><subject>Machine learning</subject><subject>Nature Conservation</subject><subject>Original Paper</subject><subject>Predictions</subject><subject>Regression analysis</subject><issn>1435-9529</issn><issn>1435-9537</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNqFkctOAyEYhSdGE2v1BVyRuB7lUmaGpWm8NGniRteEy09LnTIV6GJewOcWrdGdLoATON8Jf05VXRJ8TTBub1LZm1mNaVms5V09HlUTMmO8Fpy1xz-aitPqLKUNxoR3lEyq90XI0Pd-BSGjCCs_BNWjtNdpH50ygHYRrDe53COtElhURO-3PhephwjroQdkVVZIBYt8SYsFyUr73ucRpazMqw8rlMGsg3_bAxocgpBgqwvYg4qhPJ9XJ071CS6-z2n1cn_3PH-sl08Pi_ntsjaUdrnmRFhlCe40ECVsC442XGsLZta4VjWMOq3LbMoILRhY5ozoCDWmVY4by9m0ujrk7uJQ_pKy3Az7WEZOkuGm6ZqZYO1_Li462rHiogeXiUNKEZzcRb9VcZQEy89W5KEVWVqRX63IsUDsAKViDiuIv9F_UB9wxJR_</recordid><startdate>20240701</startdate><enddate>20240701</enddate><creator>Bai, Jun</creator><creator>Wang, Sheng</creator><creator>Xu, Qiang</creator><creator>Zhu, Junsheng</creator><creator>Li, Zhaoqi</creator><creator>Lai, Kun</creator><creator>Liu, Xingyi</creator><creator>Chen, Zongjie</creator><general>Springer Berlin Heidelberg</general><general>Springer Nature B.V</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7ST</scope><scope>7TG</scope><scope>7UA</scope><scope>8FD</scope><scope>C1K</scope><scope>F1W</scope><scope>FR3</scope><scope>H96</scope><scope>KL.</scope><scope>KR7</scope><scope>L.G</scope><scope>SOI</scope><orcidid>https://orcid.org/0000-0003-1479-0309</orcidid></search><sort><creationdate>20240701</creationdate><title>Intelligent regional subsurface prediction based on limited borehole data and interpretability stacking technique of ensemble learning</title><author>Bai, Jun ; Wang, Sheng ; Xu, Qiang ; Zhu, Junsheng ; Li, Zhaoqi ; Lai, Kun ; Liu, Xingyi ; Chen, Zongjie</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c228t-519dad108be1a9d7ef265bbdec46f7a632fbb001ac9b93ed3fc9812cc7af5cd53</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Accuracy</topic><topic>Algorithms</topic><topic>Artificial intelligence</topic><topic>Boreholes</topic><topic>Classifiers</topic><topic>Datasets</topic><topic>Decision trees</topic><topic>Dredging</topic><topic>Earth and Environmental Science</topic><topic>Earth Sciences</topic><topic>Ensemble learning</topic><topic>Excavation</topic><topic>Foundations</topic><topic>Geoecology/Natural Processes</topic><topic>Geoengineering</topic><topic>Geology</topic><topic>Geotechnical Engineering &amp; Applied Earth Sciences</topic><topic>Hydraulics</topic><topic>Learning</topic><topic>Lithology</topic><topic>Machine learning</topic><topic>Nature Conservation</topic><topic>Original Paper</topic><topic>Predictions</topic><topic>Regression analysis</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Bai, Jun</creatorcontrib><creatorcontrib>Wang, Sheng</creatorcontrib><creatorcontrib>Xu, Qiang</creatorcontrib><creatorcontrib>Zhu, Junsheng</creatorcontrib><creatorcontrib>Li, Zhaoqi</creatorcontrib><creatorcontrib>Lai, Kun</creatorcontrib><creatorcontrib>Liu, Xingyi</creatorcontrib><creatorcontrib>Chen, Zongjie</creatorcontrib><collection>CrossRef</collection><collection>Environment Abstracts</collection><collection>Meteorological &amp; Geoastrophysical Abstracts</collection><collection>Water Resources Abstracts</collection><collection>Technology Research Database</collection><collection>Environmental Sciences and Pollution Management</collection><collection>ASFA: Aquatic Sciences and Fisheries Abstracts</collection><collection>Engineering Research Database</collection><collection>Aquatic Science &amp; Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy &amp; Non-Living Resources</collection><collection>Meteorological &amp; Geoastrophysical Abstracts - Academic</collection><collection>Civil Engineering Abstracts</collection><collection>Aquatic Science &amp; Fisheries Abstracts (ASFA) Professional</collection><collection>Environment Abstracts</collection><jtitle>Bulletin of engineering geology and the environment</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Bai, Jun</au><au>Wang, Sheng</au><au>Xu, Qiang</au><au>Zhu, Junsheng</au><au>Li, Zhaoqi</au><au>Lai, Kun</au><au>Liu, Xingyi</au><au>Chen, Zongjie</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Intelligent regional subsurface prediction based on limited borehole data and interpretability stacking technique of ensemble learning</atitle><jtitle>Bulletin of engineering geology and the environment</jtitle><stitle>Bull Eng Geol Environ</stitle><date>2024-07-01</date><risdate>2024</risdate><volume>83</volume><issue>7</issue><spage>272</spage><pages>272-</pages><artnum>272</artnum><issn>1435-9529</issn><eissn>1435-9537</eissn><abstract>This study introduces an intelligent method for regional subsurface prediction using a Stacking ensemble learning approach, which incorporates K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), Gradient Boosted Decision Trees (GBDT), and Xgboost as base classifiers, with Logistic Regression (LR) serving as the meta-classifier. Leveraging data from 1119 boreholes in Zigong City, China, this method achieves a prediction accuracy of 93%, and notably improves the prediction of weak layers, with accuracy rates ranging from 71.4% to 81.5%. This enhancement is particularly significant in areas with a random distribution of excavation and backfill. Furthermore, this study employs the SHAP method (SHapley Additive explanations) to interpret the Stacking ensemble learning model, revealing that the outputs of the base classifiers enhance the feature set for the meta-classifier, effectively addressing the insensitivity of the spatial coordinates x, y, and z as input features for lithology prediction. The findings demonstrate that the expansion of effective feature dimensions is key to the superior performance of the Stacking ensemble learning method in regional subsurface lithology prediction.</abstract><cop>Berlin/Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><doi>10.1007/s10064-024-03758-y</doi><orcidid>https://orcid.org/0000-0003-1479-0309</orcidid></addata></record>
fulltext fulltext
identifier ISSN: 1435-9529
ispartof Bulletin of engineering geology and the environment, 2024-07, Vol.83 (7), p.272, Article 272
issn 1435-9529
1435-9537
language eng
recordid cdi_proquest_journals_3066864937
source SpringerLink (Online service)
subjects Accuracy
Algorithms
Artificial intelligence
Boreholes
Classifiers
Datasets
Decision trees
Dredging
Earth and Environmental Science
Earth Sciences
Ensemble learning
Excavation
Foundations
Geoecology/Natural Processes
Geoengineering
Geology
Geotechnical Engineering & Applied Earth Sciences
Hydraulics
Learning
Lithology
Machine learning
Nature Conservation
Original Paper
Predictions
Regression analysis
title Intelligent regional subsurface prediction based on limited borehole data and interpretability stacking technique of ensemble learning
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T07%3A41%3A10IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Intelligent%20regional%20subsurface%20prediction%20based%20on%20limited%20borehole%20data%20and%20interpretability%20stacking%20technique%20of%20ensemble%20learning&rft.jtitle=Bulletin%20of%20engineering%20geology%20and%20the%20environment&rft.au=Bai,%20Jun&rft.date=2024-07-01&rft.volume=83&rft.issue=7&rft.spage=272&rft.pages=272-&rft.artnum=272&rft.issn=1435-9529&rft.eissn=1435-9537&rft_id=info:doi/10.1007/s10064-024-03758-y&rft_dat=%3Cproquest_cross%3E3066598283%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3066598283&rft_id=info:pmid/&rfr_iscdi=true