Privacy-preserving Techniques Research Articles

Federated Learning : A Paradigm Shift in Collaborative Machine Learning

Federated learning (FL) has emerged as an exceptionally promising method within the realm of machine learning, enabling multiple entities to jointly train a global model while maintaining decentralized data. This paper presents a comprehensive review of federated learning methodologies, applications, and challenges. We begin by elucidating the fundamental concepts underlying FL, including federated optimization algorithms, communication protocols, and privacy-preserving techniques. Subsequently, we delve into various domains where FL has found significant traction, examples include healthcare, finance, and the Internet of Things (IoT), showcasing successful deployments and innovative strategies. Furthermore, we discuss the inherent challenges associated with federated learning, such as communication overhead, heterogeneity of data sources, and privacy concerns, and explore state- of-the-art solutions proposed in literature. Finally, we outline future research directions in federated learning, including advancements in privacy-preserving techniques, scalability improvements, and extension of FL to emerging domains. This thorough examination provides a valuable asset for researchers, practitioners, and policymakers keen on grasping the panorama of federated learning and its ramifications for collaborative machine learning in dispersed settings.

Privacy-Preserving Techniques in Generative AI and Large Language Models: A Narrative Review

Generative AI, including large language models (LLMs), has transformed the paradigm of data generation and creative content, but this progress raises critical privacy concerns, especially when models are trained on sensitive data. This review provides a comprehensive overview of privacy-preserving techniques aimed at safeguarding data privacy in generative AI, such as differential privacy (DP), federated learning (FL), homomorphic encryption (HE), and secure multi-party computation (SMPC). These techniques mitigate risks like model inversion, data leakage, and membership inference attacks, which are particularly relevant to LLMs. Additionally, the review explores emerging solutions, including privacy-enhancing technologies and post-quantum cryptography, as future directions for enhancing privacy in generative AI systems. Recognizing that achieving absolute privacy is mathematically impossible, the review emphasizes the necessity of aligning technical safeguards with legal and regulatory frameworks to ensure compliance with data protection laws. By discussing the ethical and legal implications of privacy risks in generative AI, the review underscores the need for a balanced approach that considers performance, scalability, and privacy preservation. The findings highlight the need for ongoing research and innovation to develop privacy-preserving techniques that keep pace with the scaling of generative AI, especially in large language models, while adhering to regulatory and ethical standards.

A Narrative Review of Identity, Data and Location Privacy Techniques in Edge Computing and Mobile Crowdsourcing

As digital technology advances, the proliferation of connected devices poses significant challenges and opportunities in mobile crowdsourcing and edge computing. This narrative review focuses on the need for privacy protection in these fields, emphasizing the increasing importance of data security in a data-driven world. Through an analysis of contemporary academic literature, this review provides an understanding of the current trends and privacy concerns in mobile crowdsourcing and edge computing. We present insights and highlight advancements in privacy-preserving techniques, addressing identity, data, and location privacy. This review also discusses the potential directions that can be useful resources for researchers, industry professionals, and policymakers.

Survey on Federated Learning for Intrusion Detection System: Concept, Architectures, Aggregation Strategies, Challenges, and Future Directions

Intrusion Detection Systems (IDS) are essential for securing computer networks by identifying and mitigating potential threats. However, traditional IDS face challenges related to scalability, privacy, and computational demands as network data complexity increases. Federated Learning (FL) has emerged as a promising solution, enabling collaborative model training on decentralized data sources while preserving data privacy. Each participant retains local data repositories, ensuring data sovereignty and precluding data sharing. Leveraging the FL framework, participants locally train machine learning models on their respective datasets, subsequently transmitting model updates to a central server for aggregation. The central server then disseminates the aggregated model updates to individual participants, collectively striving to bolster intrusion detection capabilities. This article presents a comprehensive survey of FL applications in IDS, covering core concepts, architectural approaches, and aggregation strategies. We evaluate the strengths and limitations of various FL methodologies for IDS, addressing privacy and security concerns and exploring privacy-preserving techniques and security protocols. Our examination of aggregation strategies within the FL framework for IDS aims to highlight their effectiveness, limitations, and potential enhancements.

Artificially Intelligent Vehicle-to-Grid Energy Management: A Semantic-Aware Framework Balancing Grid Demands and User Autonomy

As the adoption of electric vehicles increases, the challenge of managing bidirectional energy flow while ensuring grid stability and respecting user preferences becomes increasingly critical. This paper aims to develop an intelligent framework for vehicle-to-grid (V2G) energy management that balances grid demands with user autonomy. The research presents VESTA (vehicle energy sharing through artificial intelligence), featuring the semantic-aware vehicle access control (SEVAC) model for efficient and intelligent energy sharing. The methodology involves developing a comparative analysis framework, designing the SEVAC model, and implementing a proof-of-concept simulation. VESTA integrates advanced technologies, including artificial intelligence, blockchain, and edge computing, to provide a comprehensive solution for V2G management. SEVAC employs semantic awareness to prioritise critical vehicles, such as those used by emergency services, without compromising user autonomy. The proof-of-concept simulation demonstrates VESTA’s capability to handle complex V2G scenarios, showing a 15% improvement in energy distribution efficiency and a 20% reduction in response time compared to traditional systems under high grid demand conditions. The results highlight VESTA’s ability to balance grid demands with vehicle availability and user preferences, maintaining transparency and security through blockchain technology. Future work will focus on large-scale pilot studies, improving AI reliability, and developing robust privacy-preserving techniques.

RT-PPS: Real-time privacy-preserving scheme for cloud-hosted IoT data

The Internet of Things (IoT) is a rapidly growing network of devices that can communicate with each other and with cloud-based services. These devices generate vast amounts of data that can be used to provide valuable insights into user behavior, environmental conditions, and other important factors. However, as this data is collected and processed by cloud-hosted services, there is a growing concern about privacy and security. Without adequate protection, sensitive information could be exposed to hackers or other malicious actors, putting both individuals and organizations at risk. To address this challenge, real-time privacy-preserving techniques can be used to protect IoT data without compromising its value. This paper introduces an efficient Real-time privacy-preserving scheme (RT-PPS) for cloud-hosted IoT data. RT-PPS employs multi-authority attribute-based encryption on a hybrid cloud environment to keep data secure and private, while still allowing it to be processed and analyzed by cloud-hosted services. RT-PPS has efficient response time and resource consumption, which gives it the ability to handle a huge number of concurrent users at the same time without notable delay. The proposed RT-PPS has been validated through extensive experimental evaluation on a variety of configurations. Moreover, the proposed scheme has been computationally compared with the state-of-the-artwork. RT-PPS has shown excellent performance, effectiveness, and efficiency. The RT-PPS encryption time for a 1 GB dataset while considering 1024 slices is approximately 1000 ms. Also, the RT-PPS decryption time for a 1 GB ciphertext while considering 1024 slices are approximately 235 ms. Finally, RT-PPS is proven secure against any polynomial-time attacks and their variations that have at most a negligible advantage in the introduced security model. Moreover compared to most of the state-of-the-artwork, RT-PPS reduced the ciphertext size and lowered the computations in the encryption, key generation, decryption, and ciphertext update while assuring their security. By implementing RT-PPS, organizations can take advantage of the benefits of IoT data while protecting the privacy of their users and maintaining the security of their systems.

Potentials of digital twin system for analyzing travel behavior decisions

This review explores the potential of digital twin systems to provide a more holistic representation of travel behavior and support transportation planning and policymaking. The paper introduces the concept of digital twins, their key characteristics, and their applications in various domains, including transportation. It discusses the traditional methods used in travel behavior analysis and their limitations, as well as the potential advantages of digital twin systems, such as the integration of heterogeneous data sources, real-time monitoring and prediction, and the ability to simulate and evaluate various policy scenarios. The review also identifies the key components of digital twin systems, the challenges associated with their implementation, and the current state of research on digital twins and related methods in travel behavior analysis. The paper highlights research gaps and future directions, emphasizing the need for privacy-preserving techniques, real-world case studies, and the integration of digital twins with decision support systems. Finally, the review discusses the broader implications of digital twin systems for transportation planning and policymaking, concluding by emphasizing the need for interdisciplinary collaboration and stakeholder engagement to fully realize the potential of digital twins in analyzing travel behavior decisions and shaping the future of transportation systems.

Developing a Framework for Safe AI Model Development on Sensitive Healthcare Data

The Dementias Platform UK (DPUK) Data Portal holds over 60 cohort datasets from over 3 million participants with a range of multi-modal data including neuroimaging and genomics. This has meant we have seen an increasing interest in the development of AI models with the potential for clinical implementation. However, this presents a unique challenge to disclosure control when it comes to assessing these AI models for release due to the risk of attacks such as membership inference, model inversion or even just vulnerabilities in the models like overfitting. This is why we hosted a series of workshops bringing together members of the public, expert researchers, and data owners across the UK to build a framework for allowing the safe development and release of AI models trained on sensitive healthcare data. From this, we have put together recommendations and guidelines for the use of privacy-preserving techniques in AI models to protect patient privacy, and to allow the safe deployment of these models outside of trusted research environments. We also identified the unique challenges to privacy and implementation of privacy-preserving techniques in AI models regarding the use of complex data such as neuroimaging and genomics. This framework has created a path forward for supporting safe AI model development which takes into consideration rapidly evolving ethical, legal and privacy considerations.

Blockchain-based security-minded information-sharing in precast construction supply chain management with scalability, efficiency and privacy improvements

Blockchain and Interplanetary File System (IPFS) integration holds great promise for enhancing transparency and traceability in precast construction supply chain management (PCSCM). However, such integration faces challenges regarding unauthorized access to confidential data, which can lead to significant consequences, such as financial losses and legal issues. Regarding this gap, this paper proposes a hybrid privacy-preserving access control mechanism for securing blockchain-IPFS information-sharing in PCSCM with confidentiality considerations. Multiple privacy-preserving techniques (e.g., information-sharing channels, symmetric and asymmetric encryption, and lightweight proxy re-encryption) are leveraged. A prototype system demonstrates its feasibility and effectiveness, achieving demonstrably low network latency (millisecond level), efficient encryption (millisecond level), and robust data security within both blockchain and IPFS. This research contributes to a deeper understanding of blockchain-IPFS integration and provides a valuable reference point for future research and practical adoption.

GANs for Image Security Applications: A Literature Review

Generative Adversarial Networks (GANs) have earned significant attention in various domains due to their generative model’s compelling ability to generate realistic examples probably drawn from sample distribution. Image security indicates the process of protecting digital images from unauthorized access, modification, or distribution. This requires a guarantee of image privacy, integrity, and authenticity to prohibit them from being exploited by malicious attacks. GANs can also be utilized for improving image security by exploiting its generation ability in encryption, steganography, and privacy-preserving tech-niques. This paper reviews GANs-based image security techniques providing a systematic overview of current literature and comparing the role of GANs in image encryption, image steganography, and priva-cy preserving from multiple dimensions. Additionally, it outlines future research directions to further explore the potential of GANs in addressing privacy and image security concerns

VLR-BPP: An intelligent virtual location replacement based bilateral privacy-preserving architecture for edge cloud systems

Mobile Crowdsourcing (MCS) has emerged as a significant edge-cloud computing paradigm in which workers perceive data at the network edge and report it to cloud-based computing services for processing, enabling the construction of various applications. Consequently, it is imperative to achieve Bilateral Location Privacy-Preserving (BLPP) to protect the privacy of both Data Requester (DR) and workers, as disclosing location information entails many sensitive details that can result in losses for DR and workers alike. The Local Differential Privacy (LDP) approach is widely employed in Privacy-Preserving (PP) techniques due to its inherent advantages, wherein owners release data with added noise, allowing for proactive customization of privacy strength without relying on any third party. However, the current state of LDP methods presents a dilemma: when privacy protection is strong, introducing excessive location noise can lead to a decrease in the accuracy of task-worker matching, while a high rate of task-worker matching necessitates the compromise of privacy strength. In this paper, an intelligent Virtual Location Replacement based enhanced Bilateral Privacy-Preserving (VLR-BPP) architecture is proposed to improve privacy protection strength and matching accuracy in MCS simultaneously. Within the VLR-BPP architecture, a Bipartite-Graph-based Matrix Completion (BGMC) model is employed to establish the spatiotemporal correlations among data. Then, a Virtual Location Replacement (VLR) strategy is proposed to obfuscate the locations of tasks or workers to their highly correlated virtual location before publishing. Based on VLR, three preemptive location virtualization approaches are introduced: Only Task Location Virtual (OTLV), Only Workers Location Virtual (OWLV), and Both Task and Workers Location Virtual (BTWLV). For workers and DR, Randomized Response (RR) techniques and Random Matrix Multiplication Mechanism (RMM) are used to implement LDP independently. A greedy algorithm is adopted to recruit workers for tasks. In response to the data submitted by workers, BGMC imputation mechanism is utilized to enhance data quality. Finally, simulations based on real-world datasets demonstrate that the performance of our architecture surpasses existing state-of-the-art methods in privacy protection and data collection quality by 18.92∼38.17% and 15.49∼50.77%, respectively.

A Review on Data Mining Issues, Solution & Techniques

Data mining, the process of discovering patterns and knowledge from large datasets, has become a cornerstone of decision-making across various domains including business, healthcare, and finance. This paper reviews the current landscape of data mining applications, exploring the diverse techniques employed and the challenges faced. Key issues include data quality, privacy concerns, and the scalability of mining algorithms in the face of increasingly large and complex datasets. We discuss solutions to these challenges, such as advancements in data cleaning methodologies, privacy-preserving techniques, and the development of scalable algorithms. Additionally, we provide a comprehensive overview of current data mining techniques, including clustering, classification, and association rule mining, and evaluate their effectiveness in addressing real-world problems. The paper aims to provide understanding of both the potential and limitations of data mining, offering insights into future research directions and technological advancements needed to overcome existing hurdles.

Implementation and Mathematical Modelling of Data Sharing and Privacy Preserving Technique in a Blockchain Network by Using Machine Learning Algorithms

Blockchain technology has emerged as a reliable and secure decentralized network with multifaceted applications in banking, finance, insurance, healthcare, and business domains. Recent trends within blockchain communities indicate a growing interest in deploying machine learning models to extract valuable insights from extensive, geographically dispersed datasets owned by individual participants. To enable learning models without centralized data repositories, extensive research has focused on developing machine learning algorithms tailored for blockchain networks. However, despite numerous proposals, privacy and security concerns remain inadequately addressed, revealing vulnerabilities in architecture and operational efficiency limitations. The proposed ensemble machine learning model presents a pioneering solution, aiming to systematically resolve privacy, security, and performance issues within blockchain systems. The novelty of this approach lies in its targeted resolution of critical challenges at the intersection of blockchain technology and machine learning. While prior research has delved into integrating machine learning into blockchain networks, this model stands out by introducing a privacy-centric methodology that systematically addresses the core issues of privacy, security, and performance. Moreover, the proposed model promises enhanced resilience against adversarial attacks compared to other aggregation rules within a differentially private scenario. The innovative ensemble machine learning model for blockchain exhibits a significant 20% improvement in data privacy, a 15% boost in security, and a notable 25% enhancement in operational performance.

Employing Cryptographic Techniques for Data Security and Privacy Preservation

Employing Cryptographic Techniques for Data Security and Privacy Preservation

Implementing Privacy Preserving Techniques in Distributed Big Data Framework on Cloud Computing

Cloud-based deployment architectures have emerged as the preferred model for Big Data operations due to their scalability, flexibility, and cost-effectiveness. However, this shift raises new security concerns as data is no longer directly under the user's control. Securing Big Data in cloud environments is crucial for widespread adoption, but developing a comprehensive security plan is challenging without a thorough analysis of potential vulnerabilities.To address this, a novel security-by-design framework for Big Data deployment on cloud computing, termed Big Cloud, is proposed in this article. The framework leverages systematic security analysis methodologies and automated assessment tools to map domain-specific security knowledge to design best practices. Validation of the framework was conducted through an Apache Hadoop stack use case, demonstrating its effectiveness in enhancing security awareness and reducing design time.The study also evaluates the framework's strengths and limitations, identifying key challenges in the Big Cloud domain.

A secure and privacy-preserving technique based on coupled chaotic system and plaintext encryption for multimodal medical images

A secure and privacy-preserving technique based on coupled chaotic system and plaintext encryption for multimodal medical images

Privacy Protection for Blockchain-Based Healthcare IoT Systems: A Survey

To enable precision medicine and remote patient monitoring, internet of healthcare things (IoHT) has gained significant interest as a promising technique. With the widespread use of IoHT, nonetheless, privacy infringements such as IoHT data leakage have raised serious public concerns. On the other side, blockchain and distributed ledger technologies have demonstrated great potential for enhancing trustworthiness and privacy protection for IoHT systems. In this survey, a holistic review of existing blockchain-based IoHT systems is conducted to indicate the feasibility of combining blockchain and IoHT in privacy protection. In addition, various types of privacy challenges in IoHT are identified by examining general data protection regulation (GDPR). More importantly, an associated study of cutting-edge privacy-preserving techniques for the identified IoHT privacy challenges is presented. Finally, several challenges in four promising research areas for blockchain-based IoHT systems are pointed out, with the intent of motivating researchers working in these fields to develop possible solutions.

Hybrid Filtering and Probabilistic Techniques for Privacy-Preserving Community Detection in OSNs

Online Social Networks (OSNs) face the major challenge of protecting participant's privacy, due to the high dimensionality and volume of the data. In real-time social networks, where hundreds of personal details of people are shared every day, there remains a significant threat to privacy. Privacy preservation is challenging for a community detection problem due to the high computational complexity and memory requirements, especially in larger real-world OSN graphs. Although weighted nodes provide better results, as they allow capturing the frequencies of the values, the privacy preservation of sensitive attributes such as specific profiles becomes harder compared to these models. This problem can result in queries and subsequent learnings from social network profiles of specific individuals, which may be of personal, political or otherwise concern. Online social networks (OSNs) grapple with significant privacy challenges due to the extensive dimensions and vast quantities of data involved. This research fills the void in current privacy-preserving community detection methodologies, which face problems in computational complexity and memory usage in large-scale OSNs. The proposed framework seeks to bolster privacy preservation through a comprehensive multi-step process. Data filtering uses a blended data filter system to remove outliers and irrelevant data, thus enhancing the quality of the input data. Density-Oriented Clustering phase employs a density-oriented clustering model to identify communities, with each cluster representing a distinct community. Privacy Preservation component introduces a new privacy preservation technique for sensitive OSN attributes, surpassing existing k-anonymization methods. The developed density-based social network community detection model and its novel privacy-preserving scheme are evaluated using the datasets Yelp, Football, Zachary and Dolphin from the SNAP dataset. Experimental results on these datasets embed a comprehensive evaluation based on the order of each node and the graph-based networks, where each node is laden with weights as proximity values, indicating the semantic proximity between communities and individuals. The proposed framework employs the normalized mutual information (NMI), modularity (Q), Rand index and runtime measurements to demonstrate widespread advantages in the multi-dimensional functional space, including greater accuracy, cluster compatibility, and computational tractability over the existing prominent traditional models for OSNs.

A Systematic Review of Postgraduate Programmes Concerning Ethical Imperatives of Data Privacy in Sustainable Educational Data Analytics

This systematic review, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, investigated the intersection of data privacy, postgraduate educational data analytics (EDA), and sustainability. Existing literature focuses on general privacy concerns in EDA, neglecting the specific data collected and related risks in postgraduate programmes. This review addresses this gap by identifying data types used by higher education institutions in postgraduate initiatives and evaluating the adequacy of current ethical frameworks, particularly for sustainability goals. Recognising the lack of established best practices for balancing data utility and privacy, the review analyses privacy-preserving techniques. Through identifying key data types collected in postgraduate initiatives, evaluating existing ethical frameworks, and exploring privacy-preserving educational data analytics techniques, this study provided practical guidance for higher education institutions to navigate the challenges of balancing data utility and student privacy. The results suggest that higher education institutions can achieve sustainable data use by adopting a comprehensive approach that incorporates best practices, emerging technologies, and continuous monitoring to safeguard student privacy while leveraging the benefits of educational data analytics for achieving the Sustainable Development Goals.

Self-learning activation functions to increase accuracy of privacy-preserving Convolutional Neural Networks with homomorphic encryption.

The widespread adoption of cloud computing necessitates privacy-preserving techniques that allow information to be processed without disclosure. This paper proposes a method to increase the accuracy and performance of privacy-preserving Convolutional Neural Networks with Homomorphic Encryption (CNN-HE) by Self-Learning Activation Functions (SLAF). SLAFs are polynomials with trainable coefficients updated during training, together with synaptic weights, for each polynomial independently to learn task-specific and CNN-specific features. We theoretically prove its feasibility to approximate any continuous activation function to the desired error as a function of the SLAF degree. Two CNN-HE models are proposed: CNN-HE-SLAF and CNN-HE-SLAF-R. In the first model, all activation functions are replaced by SLAFs, and CNN is trained to find weights and coefficients. In the second one, CNN is trained with the original activation, then weights are fixed, activation is substituted by SLAF, and CNN is shortly re-trained to adapt SLAF coefficients. We show that such self-learning can achieve the same accuracy 99.38% as a non-polynomial ReLU over non-homomorphic CNNs and lead to an increase in accuracy (99.21%) and higher performance (6.26 times faster) than the state-of-the-art CNN-HE CryptoNets on the MNIST optical character recognition benchmark dataset.