[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120785-en":3,"doc-seo-120785-105":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},120785,34359740700684,"Finn","https://ap-avatar.wpscdn.com/avatar/1f400023980c374ae676?_k=1777273430885731487",8,"Research & Report","Proposing the Use of Hazard Analysis for Machine Learning Data Sets - Article","Data plays a central role in artificial intelligence, and machine learning model behavior is governed by the training, validation, and test datasets. Common adequacy techniques largely evaluate attribute volumes and completeness without addressing safety risks tied to missing or incorrect attributes. This paper examines existing approaches to detect too little data and improper attributes, then argues that safety-critical assurance also requires assessing hazard impact when specific attributes are absent. The authors introduce data hazard analysis as a qualitative technique to reduce GIGO risk.","Original Article  \n[https://doi.org/10.56094/jss.v58i2.253](https://doi.org/10.56094/jss.v58i2.253)  \n\n| International System Safety Society\u003Cbr> |  |  |\n| --- | --- | --- |\n|  | Journal of System Safety |  |\n|  | Established 1965 Vol. 58 No. 2 (2023) |  |\n\nProposing the Use of Hazard Analysis for Machine Learning Data Sets  \nH. Glenn Carterb, Alexander Chanb, Chris Vinegarb, Jason Rupertac   \n[a](a Corresponding author email: mailto:jason.rupert@mtsi-va.com)[ Corresponding author email: ](a Corresponding author email: mailto:jason.rupert@mtsi-va.com)[mailto:jason.rupert@mtsi-va.com](a Corresponding author email: mailto:jason.rupert@mtsi-va.com)  \nb U.S. Army Combat Capabilities Development Command Aviation & Missile Center (DEVCOM AvMC); Redstone Arsenal, AL USAc Modern Technology Solutions, Inc.; Huntsville, AL USA  \nKeywords  \nmachine learning, data assurance, data governance  \nPeer-Reviewed Gold Open Access Zero APC Fees  \nCC-BY-ND 4.0 License  \nOnline: 22-Jun-2023  \nCite As:  \nCarter, H.G. et al. Proposing the Use of Hazard Analysis for Machine Learning Data Sets. Journal of System Safety. 2023;58(2):30-39.  \n[https://doi.org/10.56094/jss.v58i](https://doi.org/10.56094/jss.v58i)[ ](https://doi.org/10.56094/jss.v58i)[2.253](2.253)  \nABSTRACT  \nThere is no debating the importance of data for artificial intelligence. The behavior of data-driven machine learning models is determined by the data set, or as the old adage states: “garbage in, garbage out (GIGO) .” While the machine learning community is still debating which techniques are necessary and sufficient to assess the adequacy of data sets, they agree some techniques are necessary. In general, most of the techniques being considered focus on evaluating the volumes of attributes. Those attributes are evaluated with respect to anticipated counts of attributes without considering the safety concerns associated with those attributes. This paper explores those techniques to identify instances of too little data and incorrect attributes. Those techniques are important; however, for safety critical applications, the assurance analyst also needs to understand the safety impact of not having specific attributes present in the machine learning data sets. To provide that information, this paper proposes a new technique the authors call data hazard analysis. The data hazard analysis provides an approach to qualitatively analyze the training data set to reduce the risk associated with the GIGO.  \nINTRODUCTION  \nThis paper focuses on a critical building block on the path to certifying machine learning software items-establishing assurance practices for the data set used to train, validate, and test the machine learning models. Key to addressing data assurance concerns associated with certifying machine learning is conducting the hazard analysis of data sets and assuring the adequacy of the data set. Thus, this paper  \nworks through what makes up data assurance for machine learning and devotes additional time on establishing hazard assessment artifacts for the data set. This paper also presents some techniques the industry is proposing for conducting data set adequacy, completeness, and representativeness, as well as an example of data hazard analysis.  \n© The Authors. Journal of System Safety is published by the International System Safety Society 30  \nOUTLINE  \nAn introduction to highlights of traditional software assurance is provided, which includes a comparison of what type of additional assurance is needed for machine learning, where data assurance plays a key role. After that introduction, what is necessary to successfully accomplish data assurance is covered, where data hazard assessment plays a foundational role. Given that foundational role, additional time is spent in this paper proposing what would be necessary for data hazard assessment. This topic is presented to the safety community to generate discussion and engagement. There are certainly additions that should be made ","cbCaieQQtfUW6g3P","https://ap.wps.com/l/cbCaieQQtfUW6g3P","pdf",1474263,1,10,"English","en",105,"# Abstract\n# Introduction\n# Background\n# Outline","[{\"question\":\"Why is hazard analysis needed for machine learning datasets in safety-critical certification?\",\"answer\":\"Because dataset adequacy techniques often focus on attribute volumes and counts, they may miss safety impacts caused by absent or incorrect attributes. Safety-critical assurance requires understanding the hazard impact of missing specific attributes.\"},{\"question\":\"What does the paper propose to address dataset-related safety risk?\",\"answer\":\"It proposes a new technique called data hazard analysis, which qualitatively analyzes the training dataset to reduce risk associated with GIGO.\"},{\"question\":\"How does the paper position hazard assessment within the overall data assurance process?\",\"answer\":\"It treats data hazard assessment as foundational to establishing assurance practices for the data used to train, validate, and test machine learning models, supporting dataset adequacy, completeness, and representativeness.\"}]","Proposing the Use of Hazard Analysis for Machine Learning Data Sets - Article | PDF",1785732027,25,{"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},"proposing-the-use-of-hazard-analysis-for-machine-learning-data-sets-article","",{"@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/proposing-the-use-of-hazard-analysis-for-machine-learning-data-sets-article/120785/",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","2026-08-03",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 hazard analysis needed for machine learning datasets in safety-critical certification?","Question",{"text":75,"@type":76},"Because dataset adequacy techniques often focus on attribute volumes and counts, they may miss safety impacts caused by absent or incorrect attributes. Safety-critical assurance requires understanding the hazard impact of missing specific attributes.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What does the paper propose to address dataset-related safety risk?",{"text":80,"@type":76},"It proposes a new technique called data hazard analysis, which qualitatively analyzes the training dataset to reduce risk associated with GIGO.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the paper position hazard assessment within the overall data assurance process?",{"text":84,"@type":76},"It treats data hazard assessment as foundational to establishing assurance practices for the data used to train, validate, and test machine learning models, supporting dataset adequacy, completeness, and representativeness.","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"]