[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123376-en":3,"doc-seo-123376-105":30,"detail-sidebar-cat-0-en-105":91},{"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":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},123376,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","Interpreting Machine Learning Pipelines Produced by Evolutionary AutoML for Biochemical Property Prediction","Machine learning models are widely used in drug discovery through QSAR-style approaches that map molecular structures to ADMET properties, yet conventional methods often lack biochemical-task customization and generalize poorly to new biochemical spaces, reducing predictive accuracy. This work introduces an evolutionary AutoML framework that builds an interpretable model for biochemical property prediction. The method integrates grammar-based genetic programming with Bayesian networks to steer search and improve pipeline interpretability. Experiments on 12 ADMET benchmarks show comparable or better predictive performance than prior methods, while the learned Bayesian network highlights which pipeline components most influence outcomes.","Kent Academic Repository  \nde Sá,Alex G.C.,Pappa,Gisele L.,Freitas,Alex A.and Ascher,D.B.(2025)Interpreting machine learning pipelines produced by evolutionary AutoML forbiochemical property prediction.In:GECCO'25 Companion:Proceedings ofthe 2025 Genetic and Evolutionary Computation Conference Companion..pp.1944-1952.ACM ISBN 979-8-4007-1464-1.  \nDownloaded from  \nhttps://kar.kent.ac.uk/110941/The University of Kent's Academic Repository KAR  \nThe version of record is available from  \nhttps://doi.org/10.1145/3712255.3734339  \nThis document version  \nAuthor's Accepted Manuscript  \nDOI for this version  \nLicence for this versionUNSPECIFIED  \nAdditional information  \nVersions of research works  \nVersions of Record  \nIf this version is the version of record,it is the same as the published version available on the publisher's web site.Cite as the published version.  \nAuthor Accepted Manuscripts  \nIf this document is identified as the Author Accepted Manuscript it is the version after peer review but before typesetting,copy editing or publisher branding.Cite as Surname,Initial.(Year)'Title of article'.To be published in Titleof Journal,Volume and issue numbers [peer-reviewed accepted version].Available at:DOl or URL(Accessed:date).  \nEnquiries  \nIf you have questions about this document contact ResearchSupport@kent.ac.uk.Please include the URL of the recordin KAR.If you believe that your,or a third party's rights have been compromised through this document please seeour Take Down policy(available from https://www.kent.ac.uk/guides/kar-the-kent-academic-repository\\#policies).  \n# Interpreting Machine Learning Pipelines Produced byEvolutionary AutoML for Biochemical Property Prediction\n\nAlex G.C.de SáGisele L.PappaBaker Heart and Diabetes InstituteComputer Science DepartmentMelbourne,Victoria,AustraliaUniversidade Federal de Minas GeraisSchool of Chemistry &Molecular BiosciencesBelo Horizonte,BrazilThe University of Queenslandglpappa@dcc.ufmg.brBrisbane City,Queensland,AustraliaAlex.deSa@baker.edu.au  \nAlex A.Freitas  \nDavid B.Ascher  \nSchool of ComputingSchool of Chemistry &Molecular BiosciencesThe University of KentThe University of QueenslandCanterbury,Kent,United KingdomBrisbane City,Queensland,Australia  \na.a.freitas@kent.ac.uk  \n## ABSTRACT\n\nMachine learning(ML)has been playing a crucial role in drugdiscovery,mainly through quantitative structure-activity relation-ship models that relate molecular structures to properties,such asabsorption,distribution,metabolism,excretion,and toxicity(AD-MET)properties.However,traditional ML approaches often lackcustomisation to a particular biochemical task and fail to generaliseto new biochemical spaces,resulting in reduced predictive perfor-mance.Automated machine learning(AutoML)has emerged toaddress these limitations by automatically selecting the suitable MLpipelines for a given input dataset.Despite its potential,AutoMLis underutilised in cheminformatics,and its decisions often lackinterpretability,reducing user trust-especially among non-expertsAccordingly,this paper proposes an evolutionary AutoML methodfor biochemical property prediction that outputs an interpretablemodel for understanding the evolved ML pipelines.It combinesgrammar-based genetic programming with Bayesian networks toguide search and enhance the searched pipelines'interpretability.The evaluation on 12 benchmark ADMET datasets showed that theproposed AutoML method obtained similar or better results thanthree existing methods.Additionally,the interpretable Bayesiannetwork identified,among the ML pipelines'components generatedby the AutoML method (i.e.components like biochemical featureextraction methods,preprocessing techniques and ML algorithms),which components affect the ML pipelines'predictive performance  \nBaker Heart and Diabetes InstituteMelbourne,Victoria,Australia  \nd.ascher@uq.edu.au  \n## CCS CONCEPTS\n\n·Computing methodologies →Machine learning;·Appliedcomputing→Bioinformatics;·Theory of computation→Evolutionar","cbCaisA0jZKHpIKN","https://ap.wps.com/l/cbCaisA0jZKHpIKN","pdf",5969133,1,10,"English","en",105,"# Abstract\n# Introduction\n## ML for ADMET and limitations\n## Evolutionary AutoML for biochemical tasks\n# Proposed method\n## Evolutionary search with grammar-based genetic programming\n## Bayesian networks for interpretability\n# Experimental evaluation\n## Benchmark datasets\n## Comparison with existing methods\n# Component analysis and interpretation\n## Feature extraction, preprocessing, and model effects\n# Conclusion","[{\"question\":\"Why is interpretability important in AutoML for biochemical property prediction?\",\"answer\":\"AutoML decisions are often hard to interpret, which can reduce user trust, especially for non-experts. This work aims to produce an interpretable model that explains evolved ML pipeline components.\"},{\"question\":\"What is the core idea of the proposed evolutionary AutoML method?\",\"answer\":\"The method evolves ML pipelines using grammar-based genetic programming, while Bayesian networks guide search and enhance interpretability of the resulting pipelines.\"},{\"question\":\"How is the approach evaluated and what are the results?\",\"answer\":\"The method is evaluated on 12 benchmark ADMET datasets. It achieves similar or better predictive performance than three existing methods, and the interpretable Bayesian network identifies which pipeline components most affect performance.\"}]","Interpreting Machine Learning Pipelines Produced by Evolutionary AutoML for Biochemical Property Prediction | PDF",1785816185,25,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"interpreting-machine-learning-pipelines-produced-by-evolutionary-automl-for-biochemical-property-prediction","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/interpreting-machine-learning-pipelines-produced-by-evolutionary-automl-for-biochemical-property-prediction/123376/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is interpretability important in AutoML for biochemical property prediction?","Question",{"text":75,"@type":76},"AutoML decisions are often hard to interpret, which can reduce user trust, especially for non-experts. This work aims to produce an interpretable model that explains evolved ML pipeline components.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is the core idea of the proposed evolutionary AutoML method?",{"text":80,"@type":76},"The method evolves ML pipelines using grammar-based genetic programming, while Bayesian networks guide search and enhance interpretability of the resulting pipelines.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the approach evaluated and what are the results?",{"text":84,"@type":76},"The method is evaluated on 12 benchmark ADMET datasets. It achieves similar or better predictive performance than three existing methods, and the interpretable Bayesian network identifies which pipeline components most affect performance.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":21,"slug":133},"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]