{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,12]],"date-time":"2026-03-12T23:44:24Z","timestamp":1773359064881,"version":"3.50.1"},"reference-count":30,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2023,8,25]],"date-time":"2023-08-25T00:00:00Z","timestamp":1692921600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,8,25]],"date-time":"2023-08-25T00:00:00Z","timestamp":1692921600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2023,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Graph neural networks (GNNs) have been employed in materials research to predict physical and functional properties, and have achieved superior performance in several application domains over prior machine learning approaches. Recent studies incorporate features of increasing complexity such as Gaussian radial functions, plane wave functions, and angular terms to augment the neural network models, with the expectation that these features are critical for achieving a high performance. Here, we propose a GNN that adopts edge convolution where hidden edge features evolve during training and extensive attention mechanisms, and operates on simple graphs with atoms as nodes and distances between them as edges. As a result, the same model can be used for very different tasks as no other domain-specific features are used. With a model that uses no feature engineering, we achieve performance comparable with state-of-the-art models with elaborate features for formation energy and band gap prediction with standard benchmarks; we achieve even better performance when the dataset size increases. Although some domain-specific datasets still require hand-crafted features to achieve state-of-the-art results, our selected architecture choices greatly reduce the need for elaborate feature engineering and still maintain predictive power in comparison.<\/jats:p>","DOI":"10.1088\/2632-2153\/acefab","type":"journal-article","created":{"date-parts":[[2023,8,11]],"date-time":"2023-08-11T23:10:16Z","timestamp":1691795416000},"page":"035030","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["Improving materials property predictions for graph neural networks with minimal feature engineering\n                  <sup>*<\/sup>"],"prefix":"10.1088","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0850-7714","authenticated-orcid":true,"given":"Guojing","family":"Cong","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3347-6983","authenticated-orcid":true,"given":"Victor","family":"Fung","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2023,8,25]]},"reference":[{"key":"mlstacefabbib1","doi-asserted-by":"publisher","first-page":"3777","DOI":"10.1039\/C6CE00407E","article-title":"A generalized method for constructing hypothetical nanoporous materials of any net topology from graph theory","volume":"18","author":"Boyd","year":"2016","journal-title":"CrystEngComm"},{"key":"mlstacefabbib2","first-page":"pp 3","article-title":"An introduction to support vector machines for data mining","author":"Burbidge","year":"2001"},{"key":"mlstacefabbib3","doi-asserted-by":"publisher","first-page":"3564","DOI":"10.1021\/acs.chemmater.9b01294","article-title":"Graph networks as a universal machine learning framework for molecules and crystals","volume":"31","author":"Chen","year":"2019","journal-title":"Chem. Mater."},{"key":"mlstacefabbib4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s43246-021-00194-3","article-title":"A geometric-information-enhanced crystal graph network for predicting properties of materials","volume":"2","author":"Cheng","year":"2021","journal-title":"Commun. Mater."},{"key":"mlstacefabbib5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41524-021-00650-1","article-title":"Atomistic line graph neural network for improved materials property predictions","volume":"7","author":"Choudhary","year":"2021","journal-title":"npj Comput. Mater."},{"key":"mlstacefabbib6","doi-asserted-by":"publisher","first-page":"5985","DOI":"10.1021\/acs.jced.9b00835","article-title":"Advances, updates and analytics for the computation-ready, experimental metal\u2013organic framework database: CoRE MOF 2019","volume":"64","author":"Chung","year":"2019","journal-title":"J. Chem. Eng. Data"},{"key":"mlstacefabbib7","first-page":"pp 658","article-title":"Extensive attention mechanisms in graph neural networks for materials discovery","author":"Cong","year":"2022"},{"key":"mlstacefabbib8","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1016\/j.ccr.2015.08.001","article-title":"Computational characterization and prediction of metal\u2013organic framework properties","volume":"307","author":"Coudert","year":"2016","journal-title":"Coord. Chem. Rev."},{"key":"mlstacefabbib9","doi-asserted-by":"publisher","first-page":"21559","DOI":"10.1021\/acsami.0c01659","article-title":"In silico discovery of covalent organic frameworks for carbon capture","volume":"12","author":"Deeg","year":"2020","journal-title":"ACS Appl. Mater. Interfaces"},{"key":"mlstacefabbib10","first-page":"pp 770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"mlstacefabbib11","doi-asserted-by":"publisher","first-page":"8066","DOI":"10.1021\/acs.chemrev.0c00004","article-title":"Big-data science in porous materials: materials genomics and machine learning","volume":"120","author":"Jablonka","year":"2020","journal-title":"Chem. Rev."},{"key":"mlstacefabbib12","doi-asserted-by":"publisher","DOI":"10.1063\/1.4812323","article-title":"The materials project: a materials genome approach to accelerating materials innovation","volume":"1","author":"Jain","year":"2013","journal-title":"APL Mater."},{"key":"mlstacefabbib13","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevMaterials.4.093801","article-title":"Orbital graph convolutional neural network for material property prediction","volume":"4","author":"Karamad","year":"2020","journal-title":"Phys. Rev. Mater."},{"key":"mlstacefabbib14","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2016"},{"key":"mlstacefabbib15","article-title":"Directional message passing for molecular graphs","author":"Klicpera","year":"2020"},{"key":"mlstacefabbib16","doi-asserted-by":"publisher","first-page":"9360","DOI":"10.1021\/acs.jpcc.0c01167","article-title":"Topological descriptors help predict guest adsorption in nanoporous materials","volume":"124","author":"Krishnapriyan","year":"2020","journal-title":"J. Phys. Chem. C"},{"key":"mlstacefabbib17","article-title":"Equiformer: equivariant graph attention transformer for 3D atomistic graphs","author":"Liao","year":"2022"},{"key":"mlstacefabbib18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-020-17755-8","article-title":"Understanding the diversity of the metal-organic framework ecosystem","volume":"11","author":"Moosavi","year":"2020","journal-title":"Nat. Commun."},{"key":"mlstacefabbib19","article-title":"Graph neural networks exponentially lose expressive power for node classification","author":"Oono","year":"2019"},{"key":"mlstacefabbib20","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevMaterials.4.063801","article-title":"Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery","volume":"4","author":"Park","year":"2020","journal-title":"Phys. Rev. Mater."},{"key":"mlstacefabbib21","article-title":"Schnet: a continuous-filter convolutional neural network for modeling quantum interactions","volume":"vol 30","author":"Sch\u00fctt","year":"2017"},{"key":"mlstacefabbib22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/ncomms13890","article-title":"Quantum-chemical insights from deep tensor neural networks","volume":"8","author":"Sch\u00fctt","year":"2017","journal-title":"Nat. Commun."},{"key":"mlstacefabbib23","article-title":"Tensor field networks: rotation-and translation-equivariant neural networks for 3D point clouds","author":"Thomas","year":"2018"},{"key":"mlstacefabbib24","doi-asserted-by":"publisher","first-page":"630","DOI":"10.1002\/jcc.26484","article-title":"porE: a code for deterministic and systematic analyses of porosities","volume":"42","author":"Trepte","year":"2021","journal-title":"J. Comput. Chem."},{"key":"mlstacefabbib25","doi-asserted-by":"publisher","first-page":"3678","DOI":"10.1021\/acs.jctc.9b00181","article-title":"Physnet: a neural network for predicting energies, forces, dipole moments and partial charges","volume":"15","author":"Unke","year":"2019","journal-title":"J. Chem. Theory Comput."},{"key":"mlstacefabbib26","doi-asserted-by":"publisher","first-page":"2041","DOI":"10.1109\/TVCG.2012.200","article-title":"Augmented topological descriptors of pore networks for material science","volume":"18","author":"Ushizima","year":"2012","journal-title":"IEEE Trans. Vis. Comput. Graph."},{"key":"mlstacefabbib27","article-title":"Attention is all you need","volume":"vol 30","author":"Vaswani","year":"2017"},{"key":"mlstacefabbib28","article-title":"Graph attention networks","author":"Velickovic","year":"2018"},{"key":"mlstacefabbib29","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.120.145301","article-title":"Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties","volume":"120","author":"Xie","year":"2018","journal-title":"Phys. Rev. Lett."},{"key":"mlstacefabbib30","first-page":"pp 15066","article-title":"Periodic graph transformers for crystal material property prediction","volume":"vol 35","author":"Yan","year":"2022"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,25]],"date-time":"2023-08-25T11:38:20Z","timestamp":1692963500000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/acefab"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,25]]},"references-count":30,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2023,8,25]]},"published-print":{"date-parts":[[2023,9,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/acefab","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,25]]},"assertion":[{"value":"Improving materials property predictions for graph neural networks with minimal feature engineering\n                  *","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2023 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2023-05-05","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2023-08-10","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2023-08-25","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}