[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127151-en":3,"doc-seo-127151-105":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},127151,3985741905716,"Rowan","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Mitigating Unfairness in Machine Learning - A Taxonomy and an Evaluation Pipeline","Big data challenges the maintenance of ethical standards in machine learning, because datasets that do not represent populations accurately can produce biased models and unfair decisions. Such risks are critical in domains like medicine and industry, where biased outcomes can cause harmful diagnoses or costly business failures. The paper addresses this by proposing a taxonomy that organizes overlapping fairness mitigation techniques and a quantitative evaluation pipeline to compare them systematically and support practitioners in selecting suitable methods.","Mitigating Unfairness in Machine Learning: A Taxonomy and an Evaluation Pipeline  \nDiscussion Paper  \nChiara Criscuolo1, * , Tommaso Dolci 1 and Mattia Salnitri1 1 Politecnico di Milano – Department of Electronics, Information and Bioengineering  \nAbstract  \nBig data poses challenges in maintaining ethical standards for reliable outcomes in machine learning. Data that inaccurately represent populations may result in biased algorithmic models, whose application leads to unfair decisions in delicate fields such as medicine and industry. To address this issue, many fairness mitigation techniques have been introduced, but the proliferation of overlapping methods complicates decision-making for data scientists. This paper proposes a taxonomy to organize these techniques and a pipeline for their evaluation, supporting practitioners in selecting the most suitable ones. The taxonomy classifies and describes techniques qualitatively, while the pipeline offers a quantitative framework for evaluation and comparison. The proposed approach supports data scientists in addressing biased models and data effectively.  \nKeywords  \nfairness, mitigation, machine learning, taxonomy, pipeline  \n1. Introduction  \nOne of the challenges of big data is assuring high ethical standards to obtain reliable and highquality results when machine learning algorithms are employed. Data that do not correctly represent the population sooner or later lead to biased machine learning models and wrong outcomes, with possible severe impacts on people and society. For example, a biased model for evidence-based medicine may lead to wrong diagnoses, or a biased model for the industry might lead to wrong business decisions. In both cases, consequences will be drastic, potentially leading to life-threatening situations in the former case and to business failures in the latter. To address this problem, fairness mitigation techniques can be used, to either modify the analyzed data or tune the model, with the goal of reducing or removing bias. Because of the importance of the issue, mitigation techniques are proliferating in the literature, frequently generating overlapping techniques that make the choice of the the data scientist extremely difficult.  \nTherefore, we propose a taxonomy to organize the fairness mitigation techniques that can be used to mitigate bias. Such a taxonomy will help data scientists navigate through the different mitigation techniques and easily identify the ones that can be applied to the specific context. To further help the selection of mitigation techniques, this paper proposes a pipeline for their  \nSEBD 2024: 32nd Symposium on Advanced Database Systems, June 23 -26, 2024, Villasimius, Sardinia, Italy * Corresponding author.  \n$ [chiara.criscuolo@polimi.it](chiara.criscuolo@polimi.it) (C. Criscuolo); [tommaso.dolci@polimi.it](tommaso.dolci@polimi.it) (T. Dolci); mattia.salnitri@polimi.it  \n(M. Salnitri)  \n © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0) .  \nevaluation, whose objective is to be used for the creation of a shared repository of evaluations that can grow incrementally and become a reference knowledge-base.  \nThe rest of the paper is structured as follows. Section 2 reports the theoretical foundations of the research work, and Section 3 describes the state of the art on taxonomies for unfairness mitigation techniques and pipelines for their evaluation. Section 4 introduces the proposed taxonomy, with an example of a structured description focused on one unfairness mitigation technique, while Section 5 illustrates the evaluation pipeline. Finally, Section 6 concludes the paper and discusses future works.  \n2. Preliminaries  \nFairness is “the absence of any prejudice or favoritism toward an individual or a group based on their inherent or acquired characteristics” [1, p.100] . It is based on the idea of protected or sensitive attribute. A protected att","cbCaioo7DctfyNrk","https://ap.wps.com/l/cbCaioo7DctfyNrk","pdf",705937,1,10,"English","en",105,"# Introduction\n# Preliminaries\n# Related Work\n# Proposed Taxonomy\n# Evaluation Pipeline\n# Conclusion","[{\"question\":\"Why does big data lead to unfair outcomes in machine learning?\",\"answer\":\"When data do not correctly represent the target population, models can learn and amplify bias, producing unfair predictions and potentially severe consequences for individuals and society.\"},{\"question\":\"What is the proposed taxonomy used for?\",\"answer\":\"The taxonomy organizes fairness mitigation techniques into clear categories so data scientists can navigate available approaches and identify which ones fit a specific context.\"},{\"question\":\"How does the evaluation pipeline help practitioners?\",\"answer\":\"The pipeline provides a quantitative framework for evaluating and comparing mitigation techniques, enabling incremental building of a shared repository of evaluation results as a reference knowledge base.\"}]","Mitigating Unfairness in Machine Learning - A Taxonomy and an Evaluation Pipeline | PDF",1785937211,25,{"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},"mitigating-unfairness-in-machine-learning-a-taxonomy-and-an-evaluation-pipeline","",{"@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/mitigating-unfairness-in-machine-learning-a-taxonomy-and-an-evaluation-pipeline/127151/",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","2026-08-05",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 does big data lead to unfair outcomes in machine learning?","Question",{"text":75,"@type":76},"When data do not correctly represent the target population, models can learn and amplify bias, producing unfair predictions and potentially severe consequences for individuals and society.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is the proposed taxonomy used for?",{"text":80,"@type":76},"The taxonomy organizes fairness mitigation techniques into clear categories so data scientists can navigate available approaches and identify which ones fit a specific context.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the evaluation pipeline help practitioners?",{"text":84,"@type":76},"The pipeline provides a quantitative framework for evaluating and comparing mitigation techniques, enabling incremental building of a shared repository of evaluation results as a reference knowledge base.","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"]