B-TBM: A Novel Deep Learning Model with Enhanced Loss Function for HAZOP Risk Classification Using Natural Language Statistical Laws

HAZOP is a paradigm of industrial safety, and the introduction of deep learning-based HAZOP text categorization marks the arrival of an intelligent era of safety analysis. However, existing risk analysis methods have limitations in processing complex texts and extracting deep risk features. To solve...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Processes 2024-10, Vol.12 (11), p.2373
Hauptverfasser: Xu, Binxin, Lu, Duhui, Gao, Dong, Zhang, Beike
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue 11
container_start_page 2373
container_title Processes
container_volume 12
creator Xu, Binxin
Lu, Duhui
Gao, Dong
Zhang, Beike
description HAZOP is a paradigm of industrial safety, and the introduction of deep learning-based HAZOP text categorization marks the arrival of an intelligent era of safety analysis. However, existing risk analysis methods have limitations in processing complex texts and extracting deep risk features. To solve this problem, this paper proposes a novel HAZOP risk event classification model based on BERT, BiLSTM, and TextCNN. The complexity of HAZOP text is revealed by introducing statistical laws of natural language, such as Zipf’s law and Heaps’ law, and the outputs of different levels of BERT are further combined linearly to collaborate with BiLSTM and TextCNN to capture long-term dependency and local contextual information for a more accurate classification task. Meanwhile, an improved loss function is proposed to effectively solve the deficiencies of the traditional cross-entropy loss function in the mislabeling process and improve the generalization ability of the model. It is experimentally demonstrated that the accuracy of the model is improved by 3% to 4% compared to the traditional BERT model in the task of severity and possibility classification of HAZOP reports. This study not only improves the accuracy and efficiency of HAZOP risk analysis, but also provides new ideas and methods for the application of natural language processing in industrial safety.
doi_str_mv 10.3390/pr12112373
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_3133379569</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3133379569</sourcerecordid><originalsourceid>FETCH-LOGICAL-c993-fde299c0c88e0319e386d323248cea794ee300e765c894017b8338e352ad09b33</originalsourceid><addsrcrecordid>eNpNUMFOwkAUbIwmEuTiF2zizaS6u492d70BgpgUMIoXL82yfYVibetuK_Huh1vERN_lvcxMZvLG884ZvQJQ9LqyjDPGQcCR1-GcC18JJo7_3adez7ktbUcxkEHY8b6G_nI4uyEDMi8_MCe3iBWJUNsiK9ZkViYttsvqDRkXG10YTEhUOkcmTWHqrCxIWloyHbwsHshj5l7JKNfOZWlm9A_77PYuc103Vuck0sW60WskT3VLu7pV7cGdO_NOUp077P3urrecjJejqR8t7u5Hg8g3SoGfJsiVMtRIiRSYQpBhAhx4XxrUQvURgVIUYWCk6lMmVhJAIgRcJ1StALrexcG2suV7g66Ot2VjizYxBgYAQgWhalWXB5Wx7acW07iy2Zu2nzGj8b7n-K9n-AbK124x</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3133379569</pqid></control><display><type>article</type><title>B-TBM: A Novel Deep Learning Model with Enhanced Loss Function for HAZOP Risk Classification Using Natural Language Statistical Laws</title><source>MDPI - Multidisciplinary Digital Publishing Institute</source><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><creator>Xu, Binxin ; Lu, Duhui ; Gao, Dong ; Zhang, Beike</creator><creatorcontrib>Xu, Binxin ; Lu, Duhui ; Gao, Dong ; Zhang, Beike</creatorcontrib><description>HAZOP is a paradigm of industrial safety, and the introduction of deep learning-based HAZOP text categorization marks the arrival of an intelligent era of safety analysis. However, existing risk analysis methods have limitations in processing complex texts and extracting deep risk features. To solve this problem, this paper proposes a novel HAZOP risk event classification model based on BERT, BiLSTM, and TextCNN. The complexity of HAZOP text is revealed by introducing statistical laws of natural language, such as Zipf’s law and Heaps’ law, and the outputs of different levels of BERT are further combined linearly to collaborate with BiLSTM and TextCNN to capture long-term dependency and local contextual information for a more accurate classification task. Meanwhile, an improved loss function is proposed to effectively solve the deficiencies of the traditional cross-entropy loss function in the mislabeling process and improve the generalization ability of the model. It is experimentally demonstrated that the accuracy of the model is improved by 3% to 4% compared to the traditional BERT model in the task of severity and possibility classification of HAZOP reports. This study not only improves the accuracy and efficiency of HAZOP risk analysis, but also provides new ideas and methods for the application of natural language processing in industrial safety.</description><identifier>ISSN: 2227-9717</identifier><identifier>EISSN: 2227-9717</identifier><identifier>DOI: 10.3390/pr12112373</identifier><language>eng</language><publisher>Basel: MDPI AG</publisher><subject>Algorithms ; Automation ; Classification ; Complexity ; Deep learning ; Efficiency ; Industrial safety ; Language ; Natural language processing ; Risk analysis ; Risk assessment ; Semantics ; Statistical models</subject><ispartof>Processes, 2024-10, Vol.12 (11), p.2373</ispartof><rights>2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c993-fde299c0c88e0319e386d323248cea794ee300e765c894017b8338e352ad09b33</cites><orcidid>0000-0002-5304-5849</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>315,782,786,27933,27934</link.rule.ids></links><search><creatorcontrib>Xu, Binxin</creatorcontrib><creatorcontrib>Lu, Duhui</creatorcontrib><creatorcontrib>Gao, Dong</creatorcontrib><creatorcontrib>Zhang, Beike</creatorcontrib><title>B-TBM: A Novel Deep Learning Model with Enhanced Loss Function for HAZOP Risk Classification Using Natural Language Statistical Laws</title><title>Processes</title><description>HAZOP is a paradigm of industrial safety, and the introduction of deep learning-based HAZOP text categorization marks the arrival of an intelligent era of safety analysis. However, existing risk analysis methods have limitations in processing complex texts and extracting deep risk features. To solve this problem, this paper proposes a novel HAZOP risk event classification model based on BERT, BiLSTM, and TextCNN. The complexity of HAZOP text is revealed by introducing statistical laws of natural language, such as Zipf’s law and Heaps’ law, and the outputs of different levels of BERT are further combined linearly to collaborate with BiLSTM and TextCNN to capture long-term dependency and local contextual information for a more accurate classification task. Meanwhile, an improved loss function is proposed to effectively solve the deficiencies of the traditional cross-entropy loss function in the mislabeling process and improve the generalization ability of the model. It is experimentally demonstrated that the accuracy of the model is improved by 3% to 4% compared to the traditional BERT model in the task of severity and possibility classification of HAZOP reports. This study not only improves the accuracy and efficiency of HAZOP risk analysis, but also provides new ideas and methods for the application of natural language processing in industrial safety.</description><subject>Algorithms</subject><subject>Automation</subject><subject>Classification</subject><subject>Complexity</subject><subject>Deep learning</subject><subject>Efficiency</subject><subject>Industrial safety</subject><subject>Language</subject><subject>Natural language processing</subject><subject>Risk analysis</subject><subject>Risk assessment</subject><subject>Semantics</subject><subject>Statistical models</subject><issn>2227-9717</issn><issn>2227-9717</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GNUQQ</sourceid><recordid>eNpNUMFOwkAUbIwmEuTiF2zizaS6u492d70BgpgUMIoXL82yfYVibetuK_Huh1vERN_lvcxMZvLG884ZvQJQ9LqyjDPGQcCR1-GcC18JJo7_3adez7ktbUcxkEHY8b6G_nI4uyEDMi8_MCe3iBWJUNsiK9ZkViYttsvqDRkXG10YTEhUOkcmTWHqrCxIWloyHbwsHshj5l7JKNfOZWlm9A_77PYuc103Vuck0sW60WskT3VLu7pV7cGdO_NOUp077P3urrecjJejqR8t7u5Hg8g3SoGfJsiVMtRIiRSYQpBhAhx4XxrUQvURgVIUYWCk6lMmVhJAIgRcJ1StALrexcG2suV7g66Ot2VjizYxBgYAQgWhalWXB5Wx7acW07iy2Zu2nzGj8b7n-K9n-AbK124x</recordid><startdate>20241029</startdate><enddate>20241029</enddate><creator>Xu, Binxin</creator><creator>Lu, Duhui</creator><creator>Gao, Dong</creator><creator>Zhang, Beike</creator><general>MDPI AG</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7SR</scope><scope>8FD</scope><scope>8FE</scope><scope>8FG</scope><scope>8FH</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BBNVY</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>BHPHI</scope><scope>CCPQU</scope><scope>D1I</scope><scope>DWQXO</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>JG9</scope><scope>KB.</scope><scope>LK8</scope><scope>M7P</scope><scope>PDBOC</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><orcidid>https://orcid.org/0000-0002-5304-5849</orcidid></search><sort><creationdate>20241029</creationdate><title>B-TBM: A Novel Deep Learning Model with Enhanced Loss Function for HAZOP Risk Classification Using Natural Language Statistical Laws</title><author>Xu, Binxin ; Lu, Duhui ; Gao, Dong ; Zhang, Beike</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c993-fde299c0c88e0319e386d323248cea794ee300e765c894017b8338e352ad09b33</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Algorithms</topic><topic>Automation</topic><topic>Classification</topic><topic>Complexity</topic><topic>Deep learning</topic><topic>Efficiency</topic><topic>Industrial safety</topic><topic>Language</topic><topic>Natural language processing</topic><topic>Risk analysis</topic><topic>Risk assessment</topic><topic>Semantics</topic><topic>Statistical models</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Xu, Binxin</creatorcontrib><creatorcontrib>Lu, Duhui</creatorcontrib><creatorcontrib>Gao, Dong</creatorcontrib><creatorcontrib>Zhang, Beike</creatorcontrib><collection>CrossRef</collection><collection>Engineered Materials Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>ProQuest Natural Science Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>Biological Science Collection</collection><collection>Proquest Central</collection><collection>Technology Collection</collection><collection>Natural Science Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Materials Science Collection</collection><collection>ProQuest Central Korea</collection><collection>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>Materials Research Database</collection><collection>Materials Science Database</collection><collection>ProQuest Biological Science Collection</collection><collection>Biological Science Database</collection><collection>Materials Science Collection</collection><collection>Access via ProQuest (Open Access)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><jtitle>Processes</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Xu, Binxin</au><au>Lu, Duhui</au><au>Gao, Dong</au><au>Zhang, Beike</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>B-TBM: A Novel Deep Learning Model with Enhanced Loss Function for HAZOP Risk Classification Using Natural Language Statistical Laws</atitle><jtitle>Processes</jtitle><date>2024-10-29</date><risdate>2024</risdate><volume>12</volume><issue>11</issue><spage>2373</spage><pages>2373-</pages><issn>2227-9717</issn><eissn>2227-9717</eissn><abstract>HAZOP is a paradigm of industrial safety, and the introduction of deep learning-based HAZOP text categorization marks the arrival of an intelligent era of safety analysis. However, existing risk analysis methods have limitations in processing complex texts and extracting deep risk features. To solve this problem, this paper proposes a novel HAZOP risk event classification model based on BERT, BiLSTM, and TextCNN. The complexity of HAZOP text is revealed by introducing statistical laws of natural language, such as Zipf’s law and Heaps’ law, and the outputs of different levels of BERT are further combined linearly to collaborate with BiLSTM and TextCNN to capture long-term dependency and local contextual information for a more accurate classification task. Meanwhile, an improved loss function is proposed to effectively solve the deficiencies of the traditional cross-entropy loss function in the mislabeling process and improve the generalization ability of the model. It is experimentally demonstrated that the accuracy of the model is improved by 3% to 4% compared to the traditional BERT model in the task of severity and possibility classification of HAZOP reports. This study not only improves the accuracy and efficiency of HAZOP risk analysis, but also provides new ideas and methods for the application of natural language processing in industrial safety.</abstract><cop>Basel</cop><pub>MDPI AG</pub><doi>10.3390/pr12112373</doi><orcidid>https://orcid.org/0000-0002-5304-5849</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 2227-9717
ispartof Processes, 2024-10, Vol.12 (11), p.2373
issn 2227-9717
2227-9717
language eng
recordid cdi_proquest_journals_3133379569
source MDPI - Multidisciplinary Digital Publishing Institute; Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals
subjects Algorithms
Automation
Classification
Complexity
Deep learning
Efficiency
Industrial safety
Language
Natural language processing
Risk analysis
Risk assessment
Semantics
Statistical models
title B-TBM: A Novel Deep Learning Model with Enhanced Loss Function for HAZOP Risk Classification Using Natural Language Statistical Laws
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-11-30T06%3A40%3A21IST&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=B-TBM:%20A%20Novel%20Deep%20Learning%20Model%20with%20Enhanced%20Loss%20Function%20for%20HAZOP%20Risk%20Classification%20Using%20Natural%20Language%20Statistical%20Laws&rft.jtitle=Processes&rft.au=Xu,%20Binxin&rft.date=2024-10-29&rft.volume=12&rft.issue=11&rft.spage=2373&rft.pages=2373-&rft.issn=2227-9717&rft.eissn=2227-9717&rft_id=info:doi/10.3390/pr12112373&rft_dat=%3Cproquest_cross%3E3133379569%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=3133379569&rft_id=info:pmid/&rfr_iscdi=true