Tuplewise Material Representation Based Machine Learning for Accurate Band Gap Prediction

The open-access material databases allowed us to approach scientific questions from a completely new perspective with machine learning methods. Here, on the basis of open-access databases, we focus on the classical band gap problem for predicting accurately the band gap of a crystalline compound usi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The journal of physical chemistry. A, Molecules, spectroscopy, kinetics, environment, & general theory Molecules, spectroscopy, kinetics, environment, & general theory, 2020-12, Vol.124 (50), p.10616-10623
Hauptverfasser: Na, Gyoung S, Jang, Seunghun, Lee, Yea-Lee, Chang, Hyunju
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 10623
container_issue 50
container_start_page 10616
container_title The journal of physical chemistry. A, Molecules, spectroscopy, kinetics, environment, & general theory
container_volume 124
creator Na, Gyoung S
Jang, Seunghun
Lee, Yea-Lee
Chang, Hyunju
description The open-access material databases allowed us to approach scientific questions from a completely new perspective with machine learning methods. Here, on the basis of open-access databases, we focus on the classical band gap problem for predicting accurately the band gap of a crystalline compound using a machine learning approach with newly developed tuplewise graph neural networks (TGNN), which is devised to automatically generate input representation of crystal structures in tuple types and to exploit crystal-level properties as one of the input features. Our method brings about a highly accurate prediction of the band gaps at hybrid functionals and GW approximation levels for multiple material data sets without heavy computational cost. Furthermore, to demonstrate the applicability of our prediction model, we provide a data set of GW band gaps for 45835 materials predicted by TGNN posing higher accuracy than standard density functional theory calculations.
doi_str_mv 10.1021/acs.jpca.0c07802
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2467847153</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2467847153</sourcerecordid><originalsourceid>FETCH-LOGICAL-a336t-47c63f33f517b7d25b1aa6dca315e5e148a600850cd230e098084585f504b3503</originalsourceid><addsrcrecordid>eNp1kMFLwzAYR4MoTqd3T5KjBzu_JE2bHufQKUwUmQdPIU2_akfX1qRF_O_N3PTmKYG894M8Qs4YTBhwdmWsn6w6ayZgIVXA98gRkxwiyZncD3dQWSQTkY3IsfcrAGCCx4dkJARXIFR2RF6XQ1fjZ-WRPpgeXWVq-oydQ49Nb_qqbei18ViEV_teNUgXaFxTNW-0bB2dWju4oAWmKejcdPTJYVHZjXdCDkpTezzdnWPycnuznN1Fi8f5_Wy6iIwQSR_FqU1EKUQpWZqnBZc5MyYprBFMokQWK5MAKAm24AIQMgUqlkqWEuJcSBBjcrHd7Vz7MaDv9bryFuvaNNgOXvM4SVWcMikCClvUutZ7h6XuXLU27ksz0JugOgTVm6B6FzQo57v1IV9j8Sf8FgzA5Rb4UdvBNeGz_-99A6GJgJs</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2467847153</pqid></control><display><type>article</type><title>Tuplewise Material Representation Based Machine Learning for Accurate Band Gap Prediction</title><source>American Chemical Society Journals</source><creator>Na, Gyoung S ; Jang, Seunghun ; Lee, Yea-Lee ; Chang, Hyunju</creator><creatorcontrib>Na, Gyoung S ; Jang, Seunghun ; Lee, Yea-Lee ; Chang, Hyunju</creatorcontrib><description>The open-access material databases allowed us to approach scientific questions from a completely new perspective with machine learning methods. Here, on the basis of open-access databases, we focus on the classical band gap problem for predicting accurately the band gap of a crystalline compound using a machine learning approach with newly developed tuplewise graph neural networks (TGNN), which is devised to automatically generate input representation of crystal structures in tuple types and to exploit crystal-level properties as one of the input features. Our method brings about a highly accurate prediction of the band gaps at hybrid functionals and GW approximation levels for multiple material data sets without heavy computational cost. Furthermore, to demonstrate the applicability of our prediction model, we provide a data set of GW band gaps for 45835 materials predicted by TGNN posing higher accuracy than standard density functional theory calculations.</description><identifier>ISSN: 1089-5639</identifier><identifier>EISSN: 1520-5215</identifier><identifier>DOI: 10.1021/acs.jpca.0c07802</identifier><identifier>PMID: 33280389</identifier><language>eng</language><publisher>United States: American Chemical Society</publisher><subject>A: New Tools and Methods in Experiment and Theory</subject><ispartof>The journal of physical chemistry. A, Molecules, spectroscopy, kinetics, environment, &amp; general theory, 2020-12, Vol.124 (50), p.10616-10623</ispartof><rights>2020 American Chemical Society</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-a336t-47c63f33f517b7d25b1aa6dca315e5e148a600850cd230e098084585f504b3503</citedby><cites>FETCH-LOGICAL-a336t-47c63f33f517b7d25b1aa6dca315e5e148a600850cd230e098084585f504b3503</cites><orcidid>0000-0003-2614-1576 ; 0000-0001-7241-5342 ; 0000-0001-8076-5211</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://pubs.acs.org/doi/pdf/10.1021/acs.jpca.0c07802$$EPDF$$P50$$Gacs$$H</linktopdf><linktohtml>$$Uhttps://pubs.acs.org/doi/10.1021/acs.jpca.0c07802$$EHTML$$P50$$Gacs$$H</linktohtml><link.rule.ids>314,780,784,2765,27076,27924,27925,56738,56788</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/33280389$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Na, Gyoung S</creatorcontrib><creatorcontrib>Jang, Seunghun</creatorcontrib><creatorcontrib>Lee, Yea-Lee</creatorcontrib><creatorcontrib>Chang, Hyunju</creatorcontrib><title>Tuplewise Material Representation Based Machine Learning for Accurate Band Gap Prediction</title><title>The journal of physical chemistry. A, Molecules, spectroscopy, kinetics, environment, &amp; general theory</title><addtitle>J. Phys. Chem. A</addtitle><description>The open-access material databases allowed us to approach scientific questions from a completely new perspective with machine learning methods. Here, on the basis of open-access databases, we focus on the classical band gap problem for predicting accurately the band gap of a crystalline compound using a machine learning approach with newly developed tuplewise graph neural networks (TGNN), which is devised to automatically generate input representation of crystal structures in tuple types and to exploit crystal-level properties as one of the input features. Our method brings about a highly accurate prediction of the band gaps at hybrid functionals and GW approximation levels for multiple material data sets without heavy computational cost. Furthermore, to demonstrate the applicability of our prediction model, we provide a data set of GW band gaps for 45835 materials predicted by TGNN posing higher accuracy than standard density functional theory calculations.</description><subject>A: New Tools and Methods in Experiment and Theory</subject><issn>1089-5639</issn><issn>1520-5215</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><recordid>eNp1kMFLwzAYR4MoTqd3T5KjBzu_JE2bHufQKUwUmQdPIU2_akfX1qRF_O_N3PTmKYG894M8Qs4YTBhwdmWsn6w6ayZgIVXA98gRkxwiyZncD3dQWSQTkY3IsfcrAGCCx4dkJARXIFR2RF6XQ1fjZ-WRPpgeXWVq-oydQ49Nb_qqbei18ViEV_teNUgXaFxTNW-0bB2dWju4oAWmKejcdPTJYVHZjXdCDkpTezzdnWPycnuznN1Fi8f5_Wy6iIwQSR_FqU1EKUQpWZqnBZc5MyYprBFMokQWK5MAKAm24AIQMgUqlkqWEuJcSBBjcrHd7Vz7MaDv9bryFuvaNNgOXvM4SVWcMikCClvUutZ7h6XuXLU27ksz0JugOgTVm6B6FzQo57v1IV9j8Sf8FgzA5Rb4UdvBNeGz_-99A6GJgJs</recordid><startdate>20201217</startdate><enddate>20201217</enddate><creator>Na, Gyoung S</creator><creator>Jang, Seunghun</creator><creator>Lee, Yea-Lee</creator><creator>Chang, Hyunju</creator><general>American Chemical Society</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0003-2614-1576</orcidid><orcidid>https://orcid.org/0000-0001-7241-5342</orcidid><orcidid>https://orcid.org/0000-0001-8076-5211</orcidid></search><sort><creationdate>20201217</creationdate><title>Tuplewise Material Representation Based Machine Learning for Accurate Band Gap Prediction</title><author>Na, Gyoung S ; Jang, Seunghun ; Lee, Yea-Lee ; Chang, Hyunju</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a336t-47c63f33f517b7d25b1aa6dca315e5e148a600850cd230e098084585f504b3503</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>A: New Tools and Methods in Experiment and Theory</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Na, Gyoung S</creatorcontrib><creatorcontrib>Jang, Seunghun</creatorcontrib><creatorcontrib>Lee, Yea-Lee</creatorcontrib><creatorcontrib>Chang, Hyunju</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>The journal of physical chemistry. A, Molecules, spectroscopy, kinetics, environment, &amp; general theory</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Na, Gyoung S</au><au>Jang, Seunghun</au><au>Lee, Yea-Lee</au><au>Chang, Hyunju</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Tuplewise Material Representation Based Machine Learning for Accurate Band Gap Prediction</atitle><jtitle>The journal of physical chemistry. A, Molecules, spectroscopy, kinetics, environment, &amp; general theory</jtitle><addtitle>J. Phys. Chem. A</addtitle><date>2020-12-17</date><risdate>2020</risdate><volume>124</volume><issue>50</issue><spage>10616</spage><epage>10623</epage><pages>10616-10623</pages><issn>1089-5639</issn><eissn>1520-5215</eissn><abstract>The open-access material databases allowed us to approach scientific questions from a completely new perspective with machine learning methods. Here, on the basis of open-access databases, we focus on the classical band gap problem for predicting accurately the band gap of a crystalline compound using a machine learning approach with newly developed tuplewise graph neural networks (TGNN), which is devised to automatically generate input representation of crystal structures in tuple types and to exploit crystal-level properties as one of the input features. Our method brings about a highly accurate prediction of the band gaps at hybrid functionals and GW approximation levels for multiple material data sets without heavy computational cost. Furthermore, to demonstrate the applicability of our prediction model, we provide a data set of GW band gaps for 45835 materials predicted by TGNN posing higher accuracy than standard density functional theory calculations.</abstract><cop>United States</cop><pub>American Chemical Society</pub><pmid>33280389</pmid><doi>10.1021/acs.jpca.0c07802</doi><tpages>8</tpages><orcidid>https://orcid.org/0000-0003-2614-1576</orcidid><orcidid>https://orcid.org/0000-0001-7241-5342</orcidid><orcidid>https://orcid.org/0000-0001-8076-5211</orcidid></addata></record>
fulltext fulltext
identifier ISSN: 1089-5639
ispartof The journal of physical chemistry. A, Molecules, spectroscopy, kinetics, environment, & general theory, 2020-12, Vol.124 (50), p.10616-10623
issn 1089-5639
1520-5215
language eng
recordid cdi_proquest_miscellaneous_2467847153
source American Chemical Society Journals
subjects A: New Tools and Methods in Experiment and Theory
title Tuplewise Material Representation Based Machine Learning for Accurate Band Gap Prediction
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T15%3A29%3A11IST&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=Tuplewise%20Material%20Representation%20Based%20Machine%20Learning%20for%20Accurate%20Band%20Gap%20Prediction&rft.jtitle=The%20journal%20of%20physical%20chemistry.%20A,%20Molecules,%20spectroscopy,%20kinetics,%20environment,%20&%20general%20theory&rft.au=Na,%20Gyoung%20S&rft.date=2020-12-17&rft.volume=124&rft.issue=50&rft.spage=10616&rft.epage=10623&rft.pages=10616-10623&rft.issn=1089-5639&rft.eissn=1520-5215&rft_id=info:doi/10.1021/acs.jpca.0c07802&rft_dat=%3Cproquest_cross%3E2467847153%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=2467847153&rft_id=info:pmid/33280389&rfr_iscdi=true