Privacy Of User Data Research Articles

SoK: Wildest Dreams: Reproducible Research in Privacy-preserving Neural Network Training

Machine Learning (ML), addresses a multitude of complex issues in multiple disciplines, including social sciences, finance, and medical research. ML models require substantial computing power and are only as powerful as the data utilized. Due to the high computational cost of ML methods, data scientists frequently use Machine Learning-as-a-Service (MLaaS) to outsource computation to external servers. However, when working with private information, like financial data or health records, outsourcing the computation might result in privacy issues. Recent advances in Privacy-Preserving Techniques (PPTs) have enabled ML training and inference over protected data through the use of Privacy-Preserving Machine Learning (PPML). However, these techniques are still at a preliminary stage and their application in real-world situations is demanding. In order to comprehend the discrepancy between theoretical research suggestions and actual applications, this work examines the past and present of PPML, focusing on Homomorphic Encryption (HE) and Secure Multi-party Computation (SMPC) applied to ML. This work primarily focuses on the ML model's training phase, where maintaining user data privacy is of utmost importance. We provide a solid theoretical background that eases the understanding of current approaches and their limitations. We also provide some preliminaries of SMPC, HE, and ML. In addition, we present a systemization of knowledge of the most recent PPML frameworks for model training and provide a comprehensive comparison in terms of the unique properties and performances on standard benchmarks. Also, we reproduce the results for some of the surveyed papers and examine at what level existing works in the field provide support for open science. We believe our work serves as a valuable contribution by raising awareness about the current gap between theoretical advancements and real-world applications in PPML, specifically regarding open-source availability, reproducibility, and usability.

Employee Satisfaction on Compensation System: Basis for an Improved Payroll Management System with Face Recognition and Attendance Monitoring Technology

Employee satisfaction with the compensation system is a critical factor influencing organizational productivity, employee retention, and overall workplace morale. Traditional payroll management systems, often plagued by inaccuracies and inefficiencies, can negatively impact employee satisfaction and trust. This study explores the integration of advanced technologies, specifically face recognition and attendance monitoring, into payroll management systems to address these challenges and enhance employee satisfaction.The research investigates the current state of employee satisfaction with existing compensation systems through a comprehensive survey conducted among employees from diverse industries. It identifies key factors contributing to dissatisfaction, such as payroll errors, delays, and perceived lack of transparency. The study then proposes an innovative payroll management system incorporating face recognition technology for accurate attendance tracking and automated payroll processing. Findings suggest that the integration of face recognition and attendance monitoring technology in payroll management systems can significantly enhance employee satisfaction by addressing common pain points and fostering a sense of fairness and transparency. By leveraging biometric data, the proposed system aims to ensure precise and real-time attendance records, thereby reducing payroll errors and administrative burdens. The face recognition technology offers a secure, contactless, and efficient method for attendance monitoring, aligning with modern workplace safety standards and employee preferences. The study concludes with recommendations for organizations considering the adoption of such technologies, emphasizing the importance of user training, data privacy, and continuous system evaluation to ensure long-term success.

Legal Regulations for Anticipating Artificial Intelligence-Based Workers through Institutional Transformation of Job Training and the Human Resources Revolution

The rapid development of artificial intelligence (AI) technology has brought great changes to the lives of mankind. The emergence of AI as part of the rapid evolution of digitas technology has led to major changes in the world's lives. This research examines the impact of AI digitalization on Indonesia's workforce and the role of regulation in mitigating negative impacts. It emphasizes the need for holistic evaluation of labor law and modernization of professional job training programs. The research results show that the use of AI in Indonesia is increasing in various sectors, including the employment sector. Moreover, data privacy protection is still inadequate to ensure data protection and privacy of AI technology users. Therefore, it is necessary to update and add regulations that regulate comprehensively, covering data protection, algorithm transparency, occupational safety and health, elimination and specific retraining related to AI, and labor rights relevant to technological developments. It is also necessary to implement new regulations that provide legal protection for the application of AI can minimize the occurrence of cybercrimes on customer data.

Research on Computer Intelligent Collaborative Filtering Algorithm for Personalized Network Data Recommendation System

In order to meet the protection needs of user privacy data in social networks, this paper proposes a computer intelligent collaborative filtering algorithm for personalized network data recommendation systems. This algorithm predicts user preferences for specific items by utilizing user evaluation information on groups of similar feature items, thereby achieving personalized recommendations. The experimental results show that as the number of project feature selections N increases, the MAE, RMSE, and NDCG5 of the algorithm gradually improve. This is mainly attributed to increasing the number of features under a fixed similarity threshold, which makes the data granularity finer and helps to describe project features more accurately. In the case of a fixed number of project feature selections N, the impact of the number of nearest neighbors s in similar groups on algorithm performance was further studied. The results showed that with the increase of s, MAE, RMSE, and NDCG5 showed a decreasing trend. Although the algorithm suffers from certain losses in recommendation accuracy, it is still within an acceptable range. It is worth noting that due to the system only using generalized data as input, user privacy data is effectively protected. Based on the comprehensive experimental results, this algorithm has significant value in practical applications.

Anomaly‐detection‐based learning for real‐time data processing in non‐intrusive load monitoring

AbstractA power system can be regarded as a cyber‐physical system with physical power networks and a cyber system based on increasing engagement with information communication technologies for smart grid functionalities for more efficient operations and control. Non‐intrusive load monitoring (NILM), an emerging smart‐grid technology, can be used to better understand the electricity usage profile and composition of smart meters using advanced data analysis algorithms. Although NILM enables various smart grid services, wider applications of NILM require addressing the challenges regarding cyber security and data privacy risks. Anomaly detection in appliance data is one of the most effective measures against potential cyber intrusions from a data perspective. This study proposes a framework of anomaly detection‐based learning algorithms to identify the anomalous periods of electricity loading data, which may be a subject for potential cyber‐attacks. Comparison studies with the hidden Markov model are performed to validate the proposed approaches. The simulation results show that these anomaly detection‐based learning algorithms work well and can precisely determine anomalous loading periods. Moreover, these trained models perform well on the testing dataset without prior knowledge of the data, providing the possibility of the real‐time assessment of power‐ loading states. The proposed framework can also be used to develop protective measures to ensure secure system operation and user data privacy.

FedCrow: Federated-Learning-Based Data Privacy Preservation in Crowd Sensing

In the process of completing large-scale and fine-grained sensing tasks for the new generation of crowd-sensing systems, the role of analysis, reasoning, and decision making based on artificial intelligence has become indispensable. Mobile crowd sensing, which is an open system reliant on the broad participation of mobile intelligent terminal devices in data sensing and computation, poses a significant risk of user privacy data leakage. To mitigate the data security threats that arise from malicious users in federated learning and the constraints of end devices in crowd-sensing applications, which are unsuitable for high computational overheads associated with traditional cryptographic security mechanisms, we propose FedCrow, which is a federated-learning-based approach for protecting crowd-sensing data that integrates federated learning with crowd sensing. FedCrow enables the training of artificial intelligence models on multiple user devices without the need to upload user data to a central server, thus mitigating the risk of crowd-sensing user data leakage. To address security vulnerabilities in the model data during the interaction process in federated learning, the system employs encryption methods suitable for crowd-sensing applications to ensure secure data transmission during the training process, thereby establishing a secure federated-learning framework for protecting crowd-sensing data. To combat potential malicious users in federated learning, a legitimate user identification method based on the user contribution level was designed using the gradient similarity principle. By filtering out malicious users, the system reduces the threat of attacks, thereby enhancing the system accuracy and security. Through various attack experiments, the system’s ability to defend against malicious user attacks was validated. The experimental results demonstrate the method’s effectiveness in countering common attacks in federated learning. Additionally, through comparative experiments, suitable encryption methods based on the size of the data in crowd-sensing applications were identified to effectively protect the data security during transmission.

Legal Regulation in the Field of Cybersecurity of Intelligent Transport (Foreign Experience)

Abstract: At present, intelligent transport is a critical part of the modernization of the transport industry around the world. An intelligent transport system combines information technology, electronic technology, data transfer technology and system engineering technology to realize intellectual features, coordinate the elements of traffic system and create an integrated management system for transport, therefore making it a subject of critical information infrastructure. In this regard, data security and its protection from cyber attacks becomes a prerequisite for ensuring reliable and efficient operation of transport systems. In fact, with the development of new energy vehicles and intelligent connected vehicles, in addition to protecting user privacy data, the security of the entire complex of smart transportation data becomes extremely important for the automotive industry, since the security concerns of their networks and data are already closely related to public safety. In this regard, many countries recognize the need for security regulations governing intelligent transport systems. The People's Republic of China is no exception in this sense. Within the country, the security system for the Internet of Vehicles (IoV) is based on regulations at various levels, which are constantly being improved as intelligent transport system develops.

Mobility-aware proactive video caching based on asynchronous federated learning in mobile edge computing systems

With the rapid advance of the Internet of Things (IoT) and the surging user-centric smart devices, video services such as short videos, real-time video streaming, and Virtual Reality (VR) gaming are emerging. Traditional cloud-centric caching is no longer able to satisfy the low-latency needs of mobile user groups. Predicting the most popular content in advance and caching it on edge devices such as small base stations (SBS) is a promising solution to alleviate network congestion and supply better Quality of Service (QoS) for mobile users. However, frequently changing popular content, limited edge caching resources, privacy of user data, and user mobility pose significant challenges to edge video caching. To tackle these challenges, we propose an efficient asynchronous federated learning (AFL)-based mobility-aware proactive video caching scheme (AF-MPVC), which significantly improves the overall network performance while preventing the risk of user data leakage. First, we model the proactive video caching problem in mobile edge computing systems and consider the impact of user mobility on AFL, representing the problem as an autoencoder (AE) model loss minimization problem. Then, a proactive video caching algorithm is proposed based on a filtering model, which uses the hidden features obtained from the trained AE model to compute the user similarity to find the popular videos that are most likely to be accessed by users and combine them with the most frequently requested videos. Meanwhile, we take into account the short duration of high mobility users staying on the current SBS and cache some videos to neighboring edge servers to have more redundant space to store more less popular videos. Finally, experiments on three real-world datasets demonstrate that the presented caching scheme outperforms other baseline caching schemes with respect to cache hit rate and latency reduction.

Anomaly detection and defense techniques in federated learning: a comprehensive review

In recent years, deep learning methods based on a large amount of data have achieved substantial success in numerous fields. However, with increases in regulations for protecting private user data, access to such data has become restricted. To overcome this limitation, federated learning (FL) has been widely utilized for training deep learning models without centralizing data. However, the inaccessibility of FL data and heterogeneity of the client data render difficulty in providing security and protecting the privacy in FL. In addition, the security and privacy anomalies in the corresponding systems significantly hinder the application of FL. Numerous studies have been proposed aiming to maintain the model security and mitigate the leakage of private training data during the FL training phase. Existing surveys categorize FL attacks from a defensive standpoint, but lack the efficiency of pinpointing attack points and implementing timely defenses. In contrast, our survey comprehensively categorizes and summarizes detected anomalies across client, server, and communication perspectives, facilitating easier identification and timely defense measures. Our survey provides an overview of the FL system and briefly introduces the FL security and privacy anomalies. Next, we detail the existing security and privacy anomalies and the methods of detection and defense from the perspectives of the client, server, and communication process. Finally, we address the security and privacy anomalies in non-independent identically distributed cases during FL and summarize the related research progress. This survey aims to provide a systematic and comprehensive review of security and privacy research in FL to help understand the progress and better apply FL in additional scenarios.

Cryptography-based location privacy protection in the Internet of Vehicles

The evolution of the Internet of Things paradigm in recent years demonstrate a significant impact on the transportation sector, leading to the emergence of a new research field, known as the Internet of Vehicles (IoV). In the IoV, vehicles can exchange information with each other and with the roadside units making use of Vehicular Ad Hoc Networks (VANETs). As this technology reaches near-to-market maturity levels, several issues arise related to the protection of users’ privacy, while the interest of adversaries for such private user data in IoV environments gets stronger. This paper aims to present a review of the state-of-the-art techniques tackling the protection of location privacy in IoV environments, as well as experimental evaluation findings regarding the usage of various cryptographic algorithms for the protection of information exchange in these networks. In the conducted evaluations, the AES algorithm has been used as the main standard, which has been coupled with several other encryption/decryption algorithms, such as RSA, ECC and NTRU. The metrics used for the evaluation include measurements over the key generation process, the certificate generation, the encryption/decryption times, the signature generation/verification times, etc. Moreover, the size of messages in the negotiation, the pseudonym exchange and the new pseudonym enabling phases has been recorded, while the energy consumption in the exchange pseudonyms phase has also been measured. All previous experiments have been carried out mainly on NS-3 and SUMO open-source software aiming to have an estimation of how the aforementioned algorithms behave under constrained resources such as CPU usage and power.

Cloud Service Authentication Based On Advanced Encryption Standard (AES) For Ensure Privacy

Considering the possible consequences of illegal access, cloud service authentication privacy must be guaranteed. Negative actors might get access to private information without the right authentication procedures, which could result in monetary losses, harm to one’s reputation, and legal repercussions. This paper proposed using the advanced encryption standard (AES) algorithm for cloud service authentication to guarantee privacy and alleviate these problems. Providing a safe and effective authentication process that ensures user data privacy is the main goal of this framework. To encrypt user credentials and other authentication related data before sending it to the cloud service provider, the proposed framework makes use of the AES algorithm. With the use of an encryption key, the data is rendered unintelligible even in the event that it is intercepted. In order to guard against man in the middle attacks and eavesdropping, the authentication system also includes secure key exchange protocols and secure communication routes. Organizations may feel secure and in control of their data stored in the cloud by implementing this cloud service authentication architecture. It provides a strong barrier against breaches of privacy, illegal access, and data breaches. A workable and efficient way to guarantee privacy and safeguard sensitive data is to employ the AES algorithm for cloud service authentication. Through the use of this framework, businesses may take use of cloud computing’s advantages while retaining control over their data and lowering the danger of illegal access.

A Blockchain-Based Privacy Preserving Intellectual Property Authentication Method

With the continuous advancement of information technology, a growing number of works, including articles, paintings, and music, are being digitized. Digital content can be swiftly shared and disseminated via the Internet. However, it is also vulnerable to malicious plagiarism, which can seriously infringe upon the rights of creators and dampen their enthusiasm. To protect creators’ rights and interests, a sophisticated method is necessary to authenticate digital intellectual property rights. Traditional authentication methods rely on centralized, trustworthy organizations that are susceptible to single points of failure. Additionally, these methods are prone to network attacks that can lead to data loss, tampering, or leakage. Moreover, the circulation of copyright information often lacks transparency and traceability in traditional systems, which leads to information asymmetry and prevents creators from controlling the use and protection of their personal information during the authentication process. Blockchain technology, with its decentralized, tamper-proof, and traceable attributes, addresses these issues perfectly. In blockchain technology, each node is a peer, ensuring the symmetry of information. However, the transparent feature of blockchains can lead to the leakage of user privacy data. Therefore, this study designs and implements an Ethereum blockchain-based intellectual property authentication scheme with privacy protection. Firstly, we propose a method that combines elliptic curve cryptography (ECC) encryption with digital signatures to achieve selective encryption of user personal information. Subsequently, an authentication algorithm based on Zero-Knowledge Succinct Non-Interactive Argument of Knowledge (zk-SNARK) is adopted to complete the authentication of intellectual property ownership while encrypting personal privacy data. Finally, we adopt the InterPlanetary File System (IPFS) to store large files, solving the problem of blockchain storage space limitations.

Current Perceptions of Virtual Reality Technology

User experience is a key predictor of future use of goods and services. The presented study collected a combination of qualitative and quantitative data from both experienced users and novices about their perceptions of virtual reality (VR) equipment, any concerns surrounding the data collected by the equipment, and facets that needed to be taken into consideration for future developments. The purpose of this research was to understand the current user experience of VR and ways in which it can be improved. The findings indicated that the majority of people have used VR, albeit infrequently, and that the most common use for it was for entertainment purposes. The most important characteristics of VR systems were judged to be the available content and price. While it was reported to be enjoyable to use, the ways in which it was suggested to be improved were through a reduction in size and weight of the headsets, and incorporating wireless capabilities. Concerns about the use of VR were the potential for sickness, discomfort, and eye strain, the disconnect from the real world and the subsequent risks that this poses, as well as the use and privacy of user data. The findings from this research can be used as a stepping stone toward the advancement of VR technologies.

Privacy‐preserving data aggregation achieving completeness of data queries in smart grid

AbstractIn smart grid systems, the control center formulates strategies and provides services by analyzing electricity consumption data. However, ensuring the privacy and security of user data is a critical concern. While traditional data aggregation schemes can provide a certain level of privacy protection for users, they also impose limitations on the control center's access to fine‐grained data. To address these challenges, we propose a privacy‐preserving data aggregation scheme supporting data query (PAQ). We designed a multi‐level data aggregation mechanism based on Paillier semi‐homomorphic encryption to achieve efficient aggregation of user data in the control center. Additionally, a data query mechanism based on electricity consumption intervals is introduced, allowing the control center to query aggregated ciphertexts for different user categories from outsourced data on the cloud server. Security analysis demonstrates that PAQ design effectively solves security issues in data aggregation and query processes. Performance analysis indicates that the proposed scheme outperforms existing solutions in terms of efficiency.

Physiological-chain: A privacy preserving physiological data sharing ecosystem

Safeguarding the privacy of physiological data has long been a focal point, especially given the increasing concern surrounding the data exchange process and its implications for privacy. In response to these challenges, we propose a two-layer blockchain-based platform comprising the Decentralized Identity Blockchain (DID Chain) and the Physiological Data Blockchain (App Chain). Within the DID chain, we deploy two contracts: the IdentityManager Contract and the PersonalIdentity Contract. Users are empowered to register their decentralized identity, while governments can leverage the IdentityManager Contract to generate decentralized identities. Once users obtain their identity, they can proceed to register the app-chain using that identity. Moreover, users can utilize the AccessControlManager chaincode within the App-chain to share de-identified information generated by their wearable devices, retaining full control over managing the access control rights of their de-identified information. Through the integration of these two chains, our system significantly enhances user data privacy, security, and trust, thereby fostering confidence in the integration of wearable devices within the home healthcare data sharing process.

Privacy Preservation of Electronic Health Records in the Modern Era: A Systematic Survey

Building a secure and privacy-preserving health data sharing framework is a topic of great interest in the healthcare sector, but its success is subject to ensuring the privacy of user data. We clarified the definitions of privacy, confidentiality and security (PCS) because these three terms have been used interchangeably in the literature. We found that researchers and developers must address the differences of these three terms when developing electronic health record (EHR) solutions. We surveyed 130 studies on EHRs, privacy-preserving techniques, and tools that were published between 2012 and 2022, aiming to preserve the privacy of EHRs. The observations and findings were summarized with the help of the identified studies framed along the survey questions addressed in the literature review. Our findings suggested that the usage of access control, blockchain, cloud-based, and cryptography techniques is common for EHR data sharing. We summarized the commonly used strategies for preserving privacy that are implemented by various EHR tools. Additionally, we collated a comprehensive list of differences and similarities between PCS. Finally, we summarized the findings in a tabular form for all EHR tools and techniques and proposed a fusion of techniques to better preserve the PCS of EHRs.

Database Access Information Security Management Simulation under Big Data Platform

When people perform database access information security management, the traditional method cannot accurately verify the identity of the visitor, the credibility of the identity information, and the security management of the access information. With the widespread application of big data technology, the amount of data in databases is rapidly increasing, which brings new challenges to information security management. The main purpose of this study is to explore how to more effectively manage the security of database access information on big data platforms. Therefore, the trusted computing platform is established to implement database access information security management under the data platform. The method determines the user behavior is credible by establishing a behavior chain of behavior based on the user identity and measuring user operation behavior. For the user's private data, the encryption/decryption module is used for security protection, preventing data from being leaked through illegal copying. A trusted metric model based on the USB Key user identity is established and a trusted platform is established. By improving the ELGamal algorithm, the IMC/IMV metrics architecture is utilized to measure platform security attributes. In the first round of anonymous authentication, the identity authentication of the platform is completely completed, and the database access information security management under the big data platform is completed. The simulation results show that in 10 experiments, the transmission time delay of TCP/IP protocol is less than 200ms, and the security of database access information is enhanced after the encryption system is established in the database. This has certain theoretical enlightenment for the improvement of database security and the optimization of information security management.

Perception of Smart Wearables Device users towards the Sustainability Concerns

Smart wearable devices have transformed the way individuals interact with technology in their daily lives, offering personalized solutions that enhance convenience, connectivity, and productivity. However, along with their benefits, concerns regarding user acceptability, accessibility, and data privacy persist. This study aims to investigate how smart wearables can improve customer experience and quality of life, addressing both opportunities and challenges they present. The theoretical framework revolves around three main pillars: personalized recommendations based on user data, seamless integration with other devices, and inclusivity and accessibility. The research examines various aspects, including user perceptions of sustainability concerns, usage patterns among young users, and possibilities for customized healthcare. Through a comprehensive literature review, gaps in research are identified, particularly regarding sustainability concerns, user awareness, and ethical considerations. The research methodology involves a quantitative survey to gauge user perceptions of sustainability issues, statistical analysis to identify trends, and theoretical exploration of sustainability factors in wearable technology. Additionally, experimental studies evaluate the feasibility and effectiveness of sustainable material substitutes. The study aims to contribute valuable insights for individuals, organizations, and policymakers to maximize the positive impact of smart wearables on users' lives while addressing sustainability and ethical concerns.

A Blockchain-Based Decentralized Public Key Infrastructure Using the Web of Trust

Internet applications rely on Secure Socket Layer (SSL)/Transport Security Layer (TSL) certifications to establish secure communication. However, the centralized nature of certificate authorities (CAs) poses a risk, as malicious third parties could exploit the CA to issue fake certificates to malicious web servers, potentially compromising the privacy and integrity of user data. In this paper, we demonstrate how the utilization of decentralized certificate verification with blockchain technology can effectively address and mitigate such attacks. We present a decentralized public key infrastructure (PKI) based on a distributed trust model, e.g., Web of Trust (WoT) and blockchain technologies, to overcome vulnerabilities like single points of failure and to prevent tampering with existing certificates. In addition, our infrastructure establishes a trusted key-ring network that decouples the authentication process from CAs in order to enhance secure certificate issuance and accelerate the revocation process. Furthermore, as a proof of concept, we present the implementation of our proposed system in the Ethereum blockchain, confirming that the proposed framework meets the five identified requirements. Our experimental results demonstrate the effectiveness of our proposed system in practice, albeit with additional overhead compared to conventional PKIs.

Federated Learning-based neural networks for airline passenger satisfaction prediction

Preserving user data privacy is paramount for airlines. However, achieving highly accurate predictions of passenger satisfaction while safeguarding the privacy of individual companys data remains a considerable challenge. To achieve this objective, we utilized a privacy-preserving machine learning technique known as Federated Learning (FL). FL allows model training on decentralized devices while ensuring data security and privacy. The FL process comprises client training processes, communication rounds, and weight aggregation. We simulate FL principles using multiple processes to enable distributed learning. Clients preprocess data and train local models, ensuring data privacy. Communication rounds involve clients downloading the global model, local training, and transmitting updates to a central server. Weight aggregation methods like Federated Averaging merge these updates, preserving data privacy. Additionally, we leverage Artificial Neural Networks (ANNs) as foundational techniques. ANNs consist of input, hidden, and output layers, with weight adjustments based on real value differences to achieve accuracy. Our approach combines FL with ANNs to demonstrate FLs potential in privacy-preserving predictive analytics. We use the Airline Passenger Satisfaction dataset for modeling and evaluate the impact of neural network depth and submodel quantity on prediction accuracy. Our experimental results reveal that neither the depth of neural networks nor the number of submodels significantly affects prediction accuracy. FL emerges as a promising approach to balance data privacy and prediction accuracy effectively.