[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-116912-en":3,"doc-seo-116912-105":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},116912,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","A Survey on Bias in Machine Learning Research","Current machine learning research on bias concentrates largely on fairness, often missing bias’s deeper roots as a systematic error introduced by people across the research lifecycle. This survey connects earlier work by proposing a taxonomy of bias sources and resulting errors in data and models, emphasizing bias in machine learning pipelines. It analyzes over forty potential pipeline bias sources with concrete examples, supporting development of methods to detect and mitigate bias for fairer, clearer, and more accurate models.","arXiv :2308 . 11254v1 [ cs .LG] 22 Aug 2023  \nA survey on bias in machine learning research  \nAgnieszka Mikolajczyk-Barelaa , Michal GrochowskiaaGda´nsk University of Technology, Gabriela Narutowicza 11/12, Gda´nsk, Poland  \nAbstract  \nCurrent research on bias in machine learning often focuses on fairness, while overlooking the roots or causes of bias. However, bias was originally defined as a ”systematic error,” often caused by humans at different stages of the research process. This article aims to bridge the gap between past literature on bias in research by providing taxonomy for potential sources of bias and errors in data and models. The paper focus on bias in machine learning pipelines. Survey analyses over forty potential sources of bias in the machine learning (ML) pipeline, providing clear examples for each. By understanding the sources and consequences of bias in machine learning, better methods can be developed for its detecting and mitigating, leading to fairer, more transparent, and more accurate ML models.  \nKeywords: Bias, deep learning, survey  \n1. Introduction  \nMachine learning has become increasingly important in various fields, from healthcare, education, administration, finance to entertainment. However, as the use of machine learning grows, so does the risk of bias in the data and models used. First of all, bias in machine learning is one of the main contributors to incorrect operation of systems, due to obvious errors in reasoning. Bias in machine learning can have serious consequences, such as perpetuating societal inequalities, discriminating against certain groups. For example, in criminal justice, biased data can lead to unfair decisions, such as disproportionately targeting certain demographics or falsely identifying someone as a criminal [7] . In healthcare, biased data can lead to inadequate diagnoses, mistreatment, or poor patient outcomes, as was the case with the skin cancer detection algorithm that showed significant bias towards lighter  \nPreprint submitted to Knowledge-Based Systems August 24, 2023  \nskinned patients [33] . As such, it is crucial to identify and address bias in machine learning to ensure fair and equitable outcomes.  \nThis survey provides an overview of the current state of knowledge regarding bias in machine learning research, including the review of biases at different stages of research and contemporary approaches to bias detection and mitigation. Bias can be broadly defined as ”a systematic deviation of results or inferences from the truth or processes leading to such deviation”(Choi et al. [51]) . In machine learning, bias is often referred to as ”a systematic error from erroneous assumptions in the learning algorithm”(Mehrabi et al. [78]) . Bias can be found in all areas of research. It can interfere with research project at any stage, including the beginning (e.g., literature review or data collection), the middle (e.g., the model training), and the end (e.g., evaluation and closure of a research project) (Choi et al. [51]) . Avoiding bias demands ongoing attention and awareness from all project members. However, it is natural that even with such efforts, errors can still occur. Nevertheless, to ensure that bias does not interfere with our research, we must first be aware of its existence.  \nTo ensure that machine learning models are not biased, researchers have developed various approaches to detect and mitigate bias in data and models. Some of these approaches include fairness metrics, debiasing techniques, and explainability methods. Fairness metrics provide quantitative measures of fairness in models by evaluating their performance across, for example, different demographic groups. Debiasing techniques aim to reduce the impact of bias in the data by removing or balancing biased features or samples. Explainability methods aim to provide insights into how the model makes its decisions, allowing for the identification and correction of any biases.  \nTo detect and mi","cbCaikSe2SwI3YSN","https://ap.wps.com/l/cbCaikSe2SwI3YSN","pdf",2402754,1,48,"English","en",105,"# Introduction\n## Bias across real-world applications\n## Definitions and scope of bias\n# Related works\n## Surveys on fairness and bias\n## Gap between prior and current research\n# Sources of bias at different stages of research\n## Taxonomy of pipeline bias sources\n## Examples and mitigation overview","[{\"question\":\"What problem does the survey aim to address in bias research?\",\"answer\":\"It addresses the gap where many studies focus on fairness while overlooking bias’s origins as a systematic error created at different stages of the research process.\"},{\"question\":\"How does the survey structure bias understanding in machine learning pipelines?\",\"answer\":\"It provides a taxonomy covering potential sources of bias and errors across the pipeline, supported by clear examples for each source.\"},{\"question\":\"What approaches are discussed for detecting and mitigating bias?\",\"answer\":\"The survey discusses fairness metrics, debiasing techniques to reduce biased features or samples, and explainability methods to identify and correct biased decision drivers.\"}]","A Survey on Bias in Machine Learning Research | PDF",1785672478,121,{"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},"a-survey-on-bias-in-machine-learning-research","",{"@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/a-survey-on-bias-in-machine-learning-research/116912/",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","2026-08-02",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},"What problem does the survey aim to address in bias research?","Question",{"text":75,"@type":76},"It addresses the gap where many studies focus on fairness while overlooking bias’s origins as a systematic error created at different stages of the research process.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the survey structure bias understanding in machine learning pipelines?",{"text":80,"@type":76},"It provides a taxonomy covering potential sources of bias and errors across the pipeline, supported by clear examples for each source.",{"name":82,"@type":73,"acceptedAnswer":83},"What approaches are discussed for detecting and mitigating bias?",{"text":84,"@type":76},"The survey discusses fairness metrics, debiasing techniques to reduce biased features or samples, and explainability methods to identify and correct biased decision drivers.","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},[93,97,101,105,110,115,120,123,128,131,135],{"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":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]