[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126191-en":3,"doc-seo-126191-105":31,"detail-sidebar-cat-0-en-105":93},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126191,549768072016,"River Wang","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Explainable artificial intelligence for machine learning prediction of bandgap energies","The bandgap governs the electrical and optical behavior of semiconductors and insulators, yet density functional theory calculations are computationally demanding and often underestimate bandgaps. Machine learning can deliver high-throughput, accurate predictions, but many models remain difficult to interpret, limiting scientific insight. This study applies explainable AI to support vector regression, gradient boosting regression, and random forest regression using permutation feature importance and multiple interpretability visualizations. The analysis highlights covalent-bond electron count and element mass density as key descriptors, revealing a dependency that bandgap decreases with increasing average mass density, supported by atomic-structure interpretation.","RESEARCH ARTICLE | NOVEMBER 04 2024  \nExplainable artificial intelligence for machine learning prediction of bandgap energies   \nTaichi Masuda  ; Katsuaki Tanabe 􀀤   \nJ. Appl. Phys. 136, 175703 (2024)  \n[https://doi.org/10.1063/5.0226151](https://doi.org/10.1063/5.0226151)  \n􀀪  \nView Online  \n􀀮  \nExport Citation  \n07 January 2025 00:51:40  \nExplainable artificial intelligence for machine learning prediction of bandgap energies   \n\n| Cite as: J. Appl. Phys. 136, 175703 (2024); doi: 10. 1063/5.0226151 Submitted: 29 June 2024 · Accepted: 18 October 2024 ·\u003Cbr>\u003Cbr>\u003Cbr>\u003Cbr>View Online Export Citation CrossMark\u003Cbr>Published Online: 4 November 2024 |\n| --- |\n| Taichi Masuda  and Katsuaki Tanabea)  |\n| AFFILIATIONS\u003Cbr>Department of Chemical Engineering, Kyoto University, Nishikyo, Kyoto 615-8510, Japan\u003Cbr>a)Author to whom correspondence should be addressed: [tanabe@cheme.kyoto-u.ac.jp](tanabe@cheme.kyoto-u.ac.jp) |\n| ABSTRACT\u003Cbr>The bandgap is an inherent property of semiconductors and insulators, significantly influencing their electrical and optical characteristics. However, theoretical calculations using the density functional theory (DFT) are time-consuming and underestimate bandgaps. Machine learning offers a promising approach for predicting bandgaps with high precision and high throughput, but its models face the difficulty of being hard to interpret. Hence, an application of explainable artificial intelligence techniques to the bandgap prediction models is necessary to enhance the model’s explainability. In our study, we analyzed the support vector regression, gradient boosting regression, and random forest regression models for reproducing the experimental and DFT bandgaps using the permutation feature importance (PFI), the partial dependence plot (PDP), the individual conditional expectation plot, and the accumulated local effects plot. Through PFI, we identified that the average number of electrons forming covalent bonds and the average mass density of the elements within compounds are particularly important features for bandgap prediction models. Furthermore, PDP visualized the dependency relationship between the characteristics of the constituent elements of compounds and the bandgap. Particularly, we revealed that there is a dependency where the bandgap decreases as the average mass density of the elements of compounds increases. This result was then theoretically interpreted based on the atomic structure. These findings provide crucial guidance for selecting promising descriptors in developing high-precision and explainable bandgap prediction models. Furthermore, this research demonstrates the utility of explainable artificial intelligence methods in the efficient exploration of potential inorganic semiconductor materials.\u003Cbr>© 2024 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license ([https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)). [https://doi.org/10.1063/5.0226151](https://doi.org/10.1063/5.0226151) |\n\nI. INTRODUCTION  \nThe bandgap is an inherent property of semiconductors and insulators, significantly influencing their electrical and optical characteristics.1 Thus, the high-precision and high-throughput calculation of bandgaps of new materials is sought after for the development of materials for optoelectronic devices, such as lightemitting diodes, photovoltaics, and scintillators.2 Furthermore, significant progress has been made in the calculation of bandgaps.3 However, the conventional density functional theory (DFT) calculations using the local density approximation4 often underestimate the bandgap by more than 30% due to the inherent defects of delocalization error and derivative discontinuity.5 To overcome these challenges and enhance the accuracy of bandgap calculations, computational methods, such as hybrid functionals6 incorporating partial nonlocal Hartree–Fock (HF) exchange and the GW7,8 ","cbCaioMUjN4bQQPt","https://ap.wps.com/l/cbCaioMUjN4bQQPt","pdf",4625351,6,1,21,"English","en",105,"# Introduction\n## Bandgap importance and computational challenges\n## Machine learning for bandgap prediction\n## Need for explainability","[{\"question\":\"Why are bandgap calculations based on DFT challenging for screening new materials?\",\"answer\":\"DFT approaches such as local-density approximation can be time-consuming and often underestimate bandgaps, making large-scale high-throughput screening difficult.\"},{\"question\":\"Which machine learning regression models are analyzed in the study?\",\"answer\":\"The work evaluates support vector regression, gradient boosting regression, and random forest regression models for reproducing experimental and DFT bandgaps.\"},{\"question\":\"What features are identified as most important for bandgap prediction?\",\"answer\":\"Permutation feature importance indicates that the average number of electrons forming covalent bonds and the average mass density of elements within compounds are particularly important.\"}]","Explainable artificial intelligence for machine learning prediction of bandgap energies | PDF",1785903715,53,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"explainable-artificial-intelligence-for-machine-learning-prediction-of-bandgap-energies","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/explainable-artificial-intelligence-for-machine-learning-prediction-of-bandgap-energies/126191/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"Why are bandgap calculations based on DFT challenging for screening new materials?","Question",{"text":77,"@type":78},"DFT approaches such as local-density approximation can be time-consuming and often underestimate bandgaps, making large-scale high-throughput screening difficult.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"Which machine learning regression models are analyzed in the study?",{"text":82,"@type":78},"The work evaluates support vector regression, gradient boosting regression, and random forest regression models for reproducing experimental and DFT bandgaps.",{"name":84,"@type":75,"acceptedAnswer":85},"What features are identified as most important for bandgap prediction?",{"text":86,"@type":78},"Permutation feature importance indicates that the average number of electrons forming covalent bonds and the average mass density of elements within compounds are particularly important.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,112,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":47,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":108,"slug":139},19,"General","general"]