[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117065-en":3,"doc-seo-117065-105":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},117065,3848291630094,"Emma Wilson","https://eur-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","Preserving data privacy in machine learning systems","The wide adoption of Machine Learning for real-world tasks increases the need to collect and process large volumes of personal and sensitive data, triggering serious data protection concerns. Privacy-enhancing technologies (PETs) are proposed to protect data and support trustworthiness required by current EU data-protection and AI rules, yet using PETs off-the-shelf is not sufficient to guarantee high-quality protection. This work systematically analyzes data-protection risks in modern ML systems from the perspective of data owners across the ML life cycle, showing that threats, risks, and PET protection levels depend on processing phase, participating roles, and deployment architecture. It further provides a framework to discuss privacy and confidentiality risks and assess privacy-preserving countermeasures, aiding compliance discussions with EU regulations and directives. It also summarizes unresolved challenges and research questions, offering a knowledge base for researchers and developers.","Computers & Security 137 (2024) 103605  \nContents lists available at ScienceDirect Computers & Security  \njournal [homepage: www.elsevier.com/locate/cose](homepage: www.elsevier.com/locate/cose)  \n| Preserving data privacy in machine learning systems Soumia Zohra El Mestari ∗ , Gabriele Lenzini, Huseyin Demirci\u003Cbr>SnT, University of Luxembourg, [2 Av. de](2 Av. de) l’Universite, Esch-sur-Alzette, L-4365, Luxembourg |  |  |  |\n| --- | --- | --- | --- |\n| A R T I C L E I N F O |  | A B S T R A C T |  |\n| Keywords:\u003Cbr>Privacy enhancing technologies Trustworthy machine learning Machine learning\u003Cbr>Differential privacy Homomorphic encryption Functional encryption\u003Cbr>Secure multiparty computation Privacy threats |  | The wide adoption of Machine Learning to solve a large set of real-life problems came with the need to collect and process large volumes of data, some of which are considered personal and sensitive, raising serious concerns about data protection. Privacy-enhancing technologies (PETs) are often indicated as a solution to protect personal data and to achieve a general trustworthiness as required by current EU regulations on data protection and AI. However, an off-the-shelf application of PETs is insufficient to ensure a high-quality of data protection, which one needs to understand. This work systematically discusses the risks against data protection in modern Machine Learning systems taking the original perspective of the data owners, who are those who hold the various data sets, data models, or both, throughout the machine learning life cycle and considering the different Machine Learning architectures. It argues that the origin of the threats, the risks against the data, and the level of protection offered by PETs depend on the data processing phase, the role of the parties involved, and the architecture where the machine learning systems are deployed. By offering a framework in which to discuss privacy and confidentiality risks for data owners and by identifying and assessing privacy-preserving countermeasures for machine learning, this work could facilitate the discussion about compliance with EU regulations and directives.\u003Cbr>We discuss current challenges and research questions that are still unsolved in the field. In this respect, this paper provides researchers and developers working on machine learning with a comprehensive body of knowledge to let them advance in the science of data protection in machine learning field as well as in closely related fields such as Artificial Intelligence. |  |\n\n1. Introduction  \nMachine Learning (ML) systems process data to learn valuable patterns that solve and improve the performance of a specific task at hand.  \nThese systems have demonstrated high performance and accuracy, which have led them to become drivers of innovation in several disciplines and sectors such as computer vision (Chai et al., 2021), autonomous transportation, health care (Ghassemi et al., 2020), biomedicine (Mamoshina et al., 2016) and law (Surden, 2014).  \nNowadays, ML systems are at the core of common technologies such as automatic handwriting, natural language recognition (Md Ali et al., 2021), speech processing (Vila et al., 2018), and biometric data analysis.  \nA performant ML system needs two critical resources: (i) massive volumes of datasets from multiple sources to represent data at various circumstances to be used in the training, and (ii) powerful computa-  \n* Principal corresponding author.  \nE-mail address: [soumia.elmestari@uni.lu](soumia.elmestari@uni.lu) (S.Z. El Mestari).  \n1 GDPR, [id. at](id. at) 2, Art 32.  \ntional resources to build the models. The use of such resources raises several technical, social, and ultimately legal demands: besides other technical requirements (e.g., robustness), ML systems are demanded tobe transparent and fair (e.g., free from bias in decision-making) and capable of protecting the data of the various parties involved, mainly data owners and model users.  \nThis latter requ","cbCaitCOGNyVg0kQ","https://ap.wps.com/l/cbCaitCOGNyVg0kQ","pdf",1490651,1,22,"English","en",105,"# Introduction\n## Machine Learning systems and data protection demands\n## Legal and organisational measures for privacy\n## Challenges in selecting and applying privacy solutions\n# Privacy-enhancing technologies and limitations\n# Threats, risks, and protection across the ML lifecycle\n# Framework for assessing privacy and confidentiality risks\n# Countermeasures for privacy-preserving machine learning\n# Open challenges and future research directions","[{\"question\":\"为什么在机器学习系统中要特别关注数据隐私与数据保护？\",\"answer\":\"机器学习需要收集并处理大量可能包含个人和敏感信息的数据。此类处理带来数据泄露风险，因此需要在现行法律与监管要求下采取适当的技术与组织措施来保护数据。\"},{\"question\":\"文中如何看待隐私增强技术（PETs）的作用与局限？\",\"answer\":\"PETs被视为保护个人数据、并满足欧盟数据保护与AI相关信任要求的解决方案。然而，直接采用现成的PET应用并不能保证高质量的数据保护，因此需要理解其带来的风险与有效性。\"},{\"question\":\"数据所有者视角下，隐私威胁与PET的保护水平与哪些因素有关？\",\"answer\":\"文中指出，威胁来源、对数据的风险以及PET提供的保护水平取决于数据处理阶段、参与方的角色，以及机器学习系统的部署架构。\"}]","Preserving data privacy in machine learning systems | PDF",1785673518,55,{"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},"preserving-data-privacy-in-machine-learning-systems","",{"@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/preserving-data-privacy-in-machine-learning-systems/117065/",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},"为什么在机器学习系统中要特别关注数据隐私与数据保护？","Question",{"text":75,"@type":76},"机器学习需要收集并处理大量可能包含个人和敏感信息的数据。此类处理带来数据泄露风险，因此需要在现行法律与监管要求下采取适当的技术与组织措施来保护数据。","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"文中如何看待隐私增强技术（PETs）的作用与局限？",{"text":80,"@type":76},"PETs被视为保护个人数据、并满足欧盟数据保护与AI相关信任要求的解决方案。然而，直接采用现成的PET应用并不能保证高质量的数据保护，因此需要理解其带来的风险与有效性。",{"name":82,"@type":73,"acceptedAnswer":83},"数据所有者视角下，隐私威胁与PET的保护水平与哪些因素有关？",{"text":84,"@type":76},"文中指出，威胁来源、对数据的风险以及PET提供的保护水平取决于数据处理阶段、参与方的角色，以及机器学习系统的部署架构。","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"]