[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119154-en":3,"doc-seo-119154-105":30,"detail-sidebar-cat-0-en-105":92},{"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":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},119154,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Evaluating Fair Feature Selection in Machine Learning for Healthcare - Fair Feature Selection Method","With the increasing adoption of machine learning in healthcare, algorithmic biases can intensify disparities across patient groups. The study evaluates fairness specifically during feature selection, noting that traditional approaches may ignore how relevance and correlations vary across subgroups. A fair feature selection method is tested by jointly optimizing a fairness metric and an error metric to balance bias and global classification error. Experiments on three public datasets show improved fairness with minimal balanced-accuracy degradation, addressing distributive and procedural fairness.","Evaluating Fair Feature Selection in Machine Learning for Healthcare  \nMd Rahat Shahriar Zawad1,2, Peter Washington1,2, PhD  \n1Department of Information and Computer Sciences, University of Hawaii at Manoa, Honolulu, HI, USA.  \n2Hawai’i Digital Health Laboratory, Honolulu, HI, USA.  \nAbstract  \nWith the universal adoption of machine learning in healthcare, the potential for the automation of societal biases to further exacerbate health disparities poses a significant risk. We explore algorithmic fairness from the perspective of feature selection. Traditional feature selection methods identify features for better decision making by removing resource-intensive, correlated, or non-relevant features but overlook how these factors may differ across subgroups. To counter these issues, we evaluate a fair feature selection method that considers equal importance to all demographic groups. We jointly considered a fairness metric and an error metric within the feature selection process to ensure a balance between minimizing both bias and global classification error. We tested our approach on three publicly available healthcare datasets. On all three datasets, we observed improvements in fairness metrics coupled with a minimal degradation of balanced accuracy. Our approach addresses both distributive and procedural fairness within the fair machine learning context.  \nIntroduction  \nAs machine learning becomes increasingly considered for clinical decision support for a broad spectrum of conditions such as cancer 1–3, cardiovascular diseases4,5, mental health disorders6–8, and infectious diseases9,10, algorithmic biases continue to be observed based on attributes such as skin color and gender 11–13. This disparity in performance often stems from historical socioeconomic and cultural biases 14, raising concerns about the adequacy of machine learning models to serve all patient groups.  \nA growing body of research has been conducted to mitigate bias in machine learning models, particularly through optimization and regularization 15–17, recalibrating models 18,19 and preprocessing of data20–22. However, research into fairness during the feature selection process, a common step in the classical machine learning pipeline, is relatively sparse by comparison. While prior studies highlight the necessity of fairness-aware frameworks in feature selection, they frequently overlook the complexities associated with the subtle interactions among differing definitions of fairness23,24 . Additionally, investigations into information-theoretic methods and kernel alignment have failed to adequately consider the impact of demographic-specific feature dependencies25,26 . Although there are proposals to assess procedural fairness through societal perceptions and legal standards27–29 and distributive fairness using genetic algorithms by aiming to balance fairness and accuracy in machine learning predictions25,26,30–32, a significant gap remains in creating an integrated, practical feature selection approach that accommodates varying definitions of fairness.  \nTo address these issues, we introduce a generalized method for addressing biases in machine learning datasets and corresponding models during the feature selection process. We analyze three different health datasets, stratified by gender, to assess and select features separately per gender. By consolidating multiple feature selection methods into a unified framework and utilizing a combined metric for final feature selection, our approach addresses distributional fairness and offers a sophisticated perspective beyond conventional procedural fairness analyses. This methodological approach presents a novel means of achieving machine learning fairness that can be combined with other more wellknown algorithmic fairness procedures to achieve a comprehensive strategy for reducing bias.  \nMethods  \nDatasets  \nWe used three distinct healthcare datasets to evaluate the applicability and functionality of our approach","cbCaicB23vqy8P2G","https://ap.wps.com/l/cbCaicB23vqy8P2G","pdf",706738,1,10,"English","en",105,"# Abstract\n# Introduction\n# Methods\n## Datasets\n## Preprocessing","[{\"question\":\"What problem does the paper address in healthcare machine learning?\",\"answer\":\"The paper addresses algorithmic bias that can arise in healthcare models and may worsen health disparities between demographic groups, especially when feature selection is performed without fairness considerations.\"},{\"question\":\"How does the proposed method incorporate fairness into feature selection?\",\"answer\":\"It evaluates feature selection by jointly considering a fairness metric and an error metric, assigning equal importance to demographic groups to balance bias reduction with classification performance.\"},{\"question\":\"What datasets and tasks were used to test the approach?\",\"answer\":\"The approach was tested on three public healthcare datasets: Tappy Keystroke (Parkinson’s Disease), Clinical and Molecular Features for Glioma Grading (LGG vs GBM), and Hospital Admission Data for Coronary Artery Disease (coronary artery disease prediction).\"}]","Evaluating Fair Feature Selection in Machine Learning for Healthcare - Fair Feature Selection Method | PDF",1785722764,25,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"evaluating-fair-feature-selection-in-machine-learning-for-healthcare-fair-feature-selection-method","",{"@graph":36,"@context":86},[37,54,69],{"@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/evaluating-fair-feature-selection-in-machine-learning-for-healthcare-fair-feature-selection-method/119154/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04","2026-08-03",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the paper address in healthcare machine learning?","Question",{"text":76,"@type":77},"The paper addresses algorithmic bias that can arise in healthcare models and may worsen health disparities between demographic groups, especially when feature selection is performed without fairness considerations.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the proposed method incorporate fairness into feature selection?",{"text":81,"@type":77},"It evaluates feature selection by jointly considering a fairness metric and an error metric, assigning equal importance to demographic groups to balance bias reduction with classification performance.",{"name":83,"@type":74,"acceptedAnswer":84},"What datasets and tasks were used to test the approach?",{"text":85,"@type":77},"The approach was tested on three public healthcare datasets: Tappy Keystroke (Parkinson’s Disease), Clinical and Molecular Features for Glioma Grading (LGG vs GBM), and Hospital Admission Data for Coronary Artery Disease (coronary artery disease prediction).","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":21,"slug":134},"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]