[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82109-en":3,"doc-seo-82109-105":29,"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":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":13,"seo_description":14,"update_tm":27,"read_time":28},82109,1099514067415,"Rowan","https://ap-avatar.wpscdn.com/avatar/100002539d78ffe74a7?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779092875211072502",8,"Research & Report","FairSelect: A Systematic Evaluation of Multi-Level and Intersectional Algorithmic Fairness","Algorithmic fairness methods help identify and mitigate bias in machine learning, but most evaluations consider single techniques and single demographic axes, leaving limited guidance for real deployments. FairSelect is a toolkit that systematically evaluates fairness mitigation strategies used alone or together across pre-processing, in-processing, and post-processing stages. It supports multiple architectures, intersectional subgroup testing, and fairness–utility comparisons across baseline, single-method, and multi-level setups. Validation uses synthetic clinical datasets and a real replication of atrial fibrillation stroke-risk prediction, showing non-additive, context-dependent mitigation effects.","FairSelect: A Systematic Evaluation of Multi-Level and Intersectional Algorithmic Fairness  \nFairSelect: Systematic, Multi-Level and Intersectional Fairness Evaluation Nick Souligne*  \nCollege of Engineering, University of Arizona, Tucson, AZ, [nasouligne@arizona.edu](nasouligne@arizona.edu)[ ](nasouligne@arizona.edu)Isabella Mixton-Garcia  \nCollege of Engineering, University of Arizona, Tucson, AZ, [irmixtongarcia@arizona.edu](irmixtongarcia@arizona.edu)[ ](irmixtongarcia@arizona.edu)Vignesh Subbian  \nCollege of Engineering, University of Arizona, Tucson, AZ, [vsubbian@arizona.edu](vsubbian@arizona.edu)  \nAlgorithmic fairness methods are increasingly used to identify and mitigate bias in machine learning models, yet most approaches are evaluated in isolation and along single demographic axes. This limits practical guidance for selecting fairness strategies, where disparities may arise across intersectional subgroups and across multiple stages of the modeling lifecycle. This work presents FairSelect, a toolkit for systematically evaluating fairness mitigation strategies applied individually and in combination across pre-processing, in-processing, and post-processing stages. FairSelect supports multiple model architectures, intersectional subgroup evaluation, and comparison of fairness–utility tradeoffs across baseline, single-method, and multi-level configurations. The framework was validated using synthetic clinical datasets designed to represent specific bias mechanisms and a real-world replication of two-year stroke risk prediction among patients with atrial fibrillation. Synthetic experiments showed that targeted fairness methods generally reduced intended subgroup disparities, while combined strategies produced larger average fairness improvements with modest utility tradeoffs. In the clinical prediction task, mitigation effects were highly variable, with some combinations improving both fairness and predictive performance while others were ineffective or counterproductive. These findings demonstrate that fairness interventions interact in non-additive and context-dependent ways. FairSelect provides a practical framework for systematically identifying fairness strategies that improve subgroup equity while preserving model performance in clinical machine learning.  \nCCS CONCEPTS • Software libraries and repositories, Modeling methodologies, Machine Learning  \nAdditional Keywords and Phrases: Algorithmic Fairness, Bias Detection, Bias Mitigation, Intersectionality  \n1 INTRODUCTION  \nThe field of algorithmic fairness has produced an extensive and quickly expanding set of methods designed to identify and mitigate biases in machine learning models. These methods span multiple stages ofthe modeling lifecycle and often target distinct sources of biases, including pre-processing approaches that modify training data distributions, in-processing techniques that incorporate fairness constraints during model optimization, and post-processing strategies that adjust model outputs after training. Without these methods, the considerable promise of predictive modeling for improving patient  \noutcomes and healthcare delivery is fundamentally limited due to the biases embedded in the data, modeling approach, or arising through other sociotechnical processes that particularly impact historically marginalized populations [1-5] . While the application of algorithmic fairness approaches offers a mechanism to address these risks, their practical application remains opaque due to diversity of biases sources, non-generalizable methods, and a lack of a systemic approach to assess best practices.  \nRecent guidance from the STANDING Together collective has reinforced the need to move beyond overall model performance evaluation towards more explicit evaluation of subgroup performance with a focus on transparent reporting of mitigation strategies [6] . These recommendations call for users of health datasets to evaluate models in contextualized groups o","cbCaitNmJD17XxAV","https://ap.wps.com/l/cbCaitNmJD17XxAV","pdf",347732,1,15,"English","en",105,"# Introduction\n## Multi-stage fairness challenges\n## Need for subgroup and intersectional evaluation\n## Motivation for multi-level method selection\n## Overview of FairSelect","[{\"question\":\"What problem does FairSelect address in algorithmic fairness evaluation?\",\"answer\":\"FairSelect targets the limitation that most fairness approaches are evaluated in isolation and along single demographic axes. It aims to provide guidance when disparities occur across intersectional subgroups and multiple stages of the modeling lifecycle.\"},{\"question\":\"How does FairSelect evaluate fairness mitigation strategies?\",\"answer\":\"FairSelect systematically tests strategies applied across pre-processing, in-processing, and post-processing stages. It supports multiple model architectures and performs intersectional subgroup evaluation while comparing fairness–utility tradeoffs across baseline, single-method, and multi-level configurations.\"},{\"question\":\"What do the synthetic and clinical replications reveal about combining fairness methods?\",\"answer\":\"Synthetic experiments generally reduce intended subgroup disparities, while combined strategies can yield larger average fairness improvements with modest utility tradeoffs. In the clinical stroke-risk prediction task, effects vary by combination—some improve both fairness and predictive performance, while others are ineffective or counterproductive, indicating non-additive interactions.\"}]",1784178259,38,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"fairselect-a-systematic-evaluation-of-multi-level-and-intersectional-algorithmic-fairness","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/fairselect-a-systematic-evaluation-of-multi-level-and-intersectional-algorithmic-fairness/82109/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does FairSelect address in algorithmic fairness evaluation?","Question",{"text":75,"@type":76},"FairSelect targets the limitation that most fairness approaches are evaluated in isolation and along single demographic axes. It aims to provide guidance when disparities occur across intersectional subgroups and multiple stages of the modeling lifecycle.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does FairSelect evaluate fairness mitigation strategies?",{"text":80,"@type":76},"FairSelect systematically tests strategies applied across pre-processing, in-processing, and post-processing stages. It supports multiple model architectures and performs intersectional subgroup evaluation while comparing fairness–utility tradeoffs across baseline, single-method, and multi-level configurations.",{"name":82,"@type":73,"acceptedAnswer":83},"What do the synthetic and clinical replications reveal about combining fairness methods?",{"text":84,"@type":76},"Synthetic experiments generally reduce intended subgroup disparities, while combined strategies can yield larger average fairness improvements with modest utility tradeoffs. In the clinical stroke-risk prediction task, effects vary by combination—some improve both fairness and predictive performance, while others are ineffective or counterproductive, indicating non-additive interactions.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]