[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119017-en":3,"doc-seo-119017-105":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},119017,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","A Survey of Data Security - Practices from Cybersecurity and Challenges of Machine Learning","Machine learning (ML) is increasingly deployed in critical systems, making the protection of training and testing data essential. This survey explains how established cybersecurity practices do not directly transfer to ML-enabled pipelines, which introduce new attack vectors and additional security and privacy risks across data collection, preprocessing, training, testing, and deployment. It consolidates shared foundations spanning cryptography, access control, zero trust architectures, homomorphic encryption, differential privacy, and federated learning to support cross-domain discussion.","A Survey of Data Security: Practices from Cybersecurity and  \nChallenges of Machine Learning  \narXiv :2310 .045 13v 3 [ cs .CR] 4 Dec 2023  \nPadmaksha Roy Virginia Tech  \nVirginia, USA  \nJaganmohan Chandrasekaran  \nNational Security Institute Virginia Tech  \nVirginia, USA  \nErin Lanus∗ National Security Institute Virginia Tech  \nVirginia, USA  \nLaura Freeman  \nNational Security Institute Virginia Tech  \nVirginia, USA  \nJeremy Werner  \nDirector Operational Test and Evaluation (DOT&E) USA  \nABSTRACT  \nMachine learning (ML) is increasingly being deployed in critical systems. The data dependence of ML makes securing data used to train and test ML-enabled systems of utmost importance. While the field of cybersecurity has well-established practices for securing information, ML-enabled systems create new attack vectors. Furthermore, data science and cybersecurity domains adhere to their own set ofskills and terminologies. This survey aims to present background information for experts in both domains in topics such as cryptography, access control, zero trust architectures, homomorphic encryption, differential privacy for machine learning, and federated learning to establish shared foundations and promote advancements in data security.  \nKEYWORDS  \ndata security, data privacy, federated learning, private machine learning  \n1 INTRODUCTION  \nWith the abundance of data in the digital age, organizations are finding new ways to create, store and transfer data at an evergrowing pace which has called for best practices in data governance to be formalized. IBM defines data security as “the practice of protecting digital information from unauthorized access, corruption, or theft throughout its entire life cycle” [1] . Data is an essential component of information systems, and data security issues need to be considered at each level of the system. That is, data resides on physical devices, is manipulated by software processes, and is foundational for most applications in autonomy andartificial intelligence (AI), such as public health and autonomous vehicles. Thus, data security is informed by many aspects of cybersecurity, ranging from physical security of data storage, network security over which data travels, cryptographic algorithms to encrypt data and cryptographic protocols to achieve authentication, logical aspects of access control mechanisms, to social, legal, and administrative policies for data use and governance.  \n∗Corresponding author [lanus@vt.edu](lanus@vt.edu)  \nA subfield ofAI, machine learning (ML) is particularly data dependent and is increasingly included as a component of information systems with many opportunities along the ML lifecycle – ranging across data collection, preprocessing, training, testing, and deployment – for data to be adversarially or accidentally compromised. In addition to incorporating established cybersecurity best practices for protecting information in data management platforms, new challenges to security and privacy presented by ML pipelines need to be addressed. The goal of this survey is to inform the data scientist and ML practitioner of existing cybersecurity mechanisms that are relevant to data security and to highlight security and privacy concerns presented by the deployment of ML for the cybersecurity expert in order to establish a common foundation for discussion between these two groups and promote advancementsin data security.  \n2 CYBERSECURITY APPROACHES TO DATA SECURITY  \n2.1 Security and Privacy Concepts  \nThe cybersecurity of an information system is most commonly discussed in terms of three foundational security properties – confidentiality, integrity, and availability – denoted the “CIA triad.” In this context, confidentiality is defined as “preserving authorized restrictions on information access and disclosure including means for protecting personal privacy and proprietary information” [2], ensuring that “sensitive information is not disclosed to unauthorized entities.” Integrity guards aga","cbCaipw7qtmxFHuB","https://ap.wps.com/l/cbCaipw7qtmxFHuB","pdf",344382,1,18,"English","en",105,"# Abstract\n# Introduction\n# Cybersecurity Approaches to Data Security\n## Security and Privacy Concepts\n## Data Governance and Security Considerations\n# Machine Learning Lifecycle Security Challenges\n## Data Collection and Preprocessing Risks\n## Training, Testing, and Deployment Threats\n# Cryptography and Privacy-Preserving Techniques\n## Homomorphic Encryption and Access Control\n## Differential Privacy and Federated Learning","[{\"question\":\"Why does data security matter specifically for machine learning systems?\",\"answer\":\"ML systems depend heavily on data across the lifecycle, so both adversarial and accidental compromise can occur during collection, preprocessing, training, testing, and deployment. Protecting training and test data is therefore crucial for secure ML operation.\"},{\"question\":\"What are the three core properties used to discuss cybersecurity in the survey?\",\"answer\":\"The survey uses the CIA triad: confidentiality, integrity, and availability. Each property is connected to specific mechanisms such as cryptographic techniques, access control, communication/network protections, and redundancy of storage.\"},{\"question\":\"Which privacy- and security-preserving techniques does the survey highlight for ML?\",\"answer\":\"It focuses on foundations including cryptography, access control, zero trust architectures, homomorphic encryption, differential privacy for machine learning, and federated learning to address security and privacy concerns.\"}]","A Survey of Data Security - Practices from Cybersecurity and Challenges of Machine Learning | PDF",1785721930,45,{"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-of-data-security-practices-from-cybersecurity-and-challenges-of-machine-learning","",{"@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-of-data-security-practices-from-cybersecurity-and-challenges-of-machine-learning/119017/",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 does data security matter specifically for machine learning systems?","Question",{"text":75,"@type":76},"ML systems depend heavily on data across the lifecycle, so both adversarial and accidental compromise can occur during collection, preprocessing, training, testing, and deployment. Protecting training and test data is therefore crucial for secure ML operation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What are the three core properties used to discuss cybersecurity in the survey?",{"text":80,"@type":76},"The survey uses the CIA triad: confidentiality, integrity, and availability. Each property is connected to specific mechanisms such as cryptographic techniques, access control, communication/network protections, and redundancy of storage.",{"name":82,"@type":73,"acceptedAnswer":83},"Which privacy- and security-preserving techniques does the survey highlight for ML?",{"text":84,"@type":76},"It focuses on foundations including cryptography, access control, zero trust architectures, homomorphic encryption, differential privacy for machine learning, and federated learning to address security and privacy concerns.","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"]