Privacy Cost Research Articles

The Cost of Privacy: Welfare Effects of the Disclosure of COVID-19 Cases

Abstract South Korea publicly disclosed detailed location information of individuals who tested positive for COVID-19. We quantify the effect of public disclosure on the transmission of the virus and economic losses in Seoul. The change in commuting patterns due to public disclosure lowers the number of cases by 60,000 and the number of deaths by 2,000 in Seoul over two years. Compared to a city-wide lockdown that results in the same number of cases over two years as the disclosure scenario, the economic cost of such a lockdown is almost four times higher.

Federated Learning Framework with Straggling Mitigation and Privacy-Awareness for AI-based Mobile Application Services

This work proposes a novel framework to address straggling and privacy issues for federated learning (FL)-based mobile application services, considering limited computing/communications resources at mobile users (MUs)/mobile application provider (MAP), privacy cost, the rationality and incentive competition among MUs in contributing data to the MAP. Particularly, the MAP first determines a set of the best MUs for the FL process based on MUs' provided information/features. Then, each selected MU can encrypt part of local data and upload the encrypted data to the MAP for an encrypted training process, in addition to the local training process. For that, the selected MU can propose a contract to the MAP according to its expected local and encrypted data. To find optimal contracts that can maximize utilities while maintaining high learning quality of the system, we develop a multi-principal one-agent contract-based problem considering the MUs' privacy cost, the MAP's limited computing resources, and asymmetric information between the MAP and MUs. Experiments with a real-world dataset show that our framework can speed up training time up to 49% and improve prediction accuracy up to 4.6 times while enhancing network's social welfare up to 114% under the privacy cost consideration compared with those of baseline methods.

Privacy Costs and Consumer Data Acquisition: An Economic Analysis of Data Privacy Regulation

Privacy Costs and Consumer Data Acquisition: An Economic Analysis of Data Privacy Regulation

Towards Personalized Privacy-Preserving Incentive for Truth Discovery in Mobile Crowdsensing Systems

Incentive mechanisms are essential for stimulating adequate worker participation to achieve good truth discovery performance in mobile crowdsensing (MCS) systems. However, most of existing incentive mechanisms only consider compensating workers’ sensing cost, while the cost incurred by potential privacy leakage has been largely neglected. Moreover, none of existing privacy-preserving incentive mechanisms has incorporated workers’ different privacy preferences to provide personalized payments for them. In this paper, we propose a contract-based personalized privacy-preserving incentive mechanism for truth discovery in MCS systems, named Paris-TD, which provides personalized payments for workers as a compensation for privacy cost while achieving accurate truth discovery. The basic idea is that the platform offers a set of different contracts to workers with different privacy preferences, and each worker chooses to sign a contract which specifies a privacy-preserving degree (PPD) and the corresponding payment the worker will receive if she submits perturbed data with that PPD. Specifically, we respectively design a set of optimal contracts analytically under both full and incomplete information models, which maximize the truth discovery accuracy under a given budget, while satisfying the individual rationality and incentive compatibility properties. The feasibility and effectiveness of Paris-TD are validated through experiments on both synthetic and real-world datasets.

Human Gait Data Augmentation and Trajectory Prediction for Lower-Limb Rehabilitation Robot Control Using GANs and Attention Mechanism

To date, several alterations in the gait pattern can be treated through rehabilitative approaches and robot assisted therapy (RAT). Gait data and gait trajectories are essential in specific exoskeleton control strategies. Nevertheless, the scarcity of human gait data due to the high cost of data collection or privacy concerns can hinder the performance of controllers or models. This paper thus first creates a GANs-based (Generative Adversarial Networks) data augmentation method to generate synthetic human gait data while still retaining the dynamics of the real gait data. Then, both the real collected and the synthesized gait data are fed to our constructed two-stage attention model for gait trajectories prediction. The real human gait data are collected with the five healthy subjects recruited from an optical motion capture platform. Experimental results indicate that the created GANs-based data augmentation model can synthesize realistic-looking multi-dimensional human gait data. Also, the two-stage attention model performs better compared with the LSTM model; the attention mechanism shows a higher capacity of learning dependencies between the historical gait data to accurately predict the current values of the hip joint angles and knee joint angles in the gait trajectory. The predicted gait trajectories depending on the historical gait data can be further used for gait trajectory tracking strategies.

Ulixes: Facial Recognition Privacy with Adversarial Machine Learning

Abstract Facial recognition tools are becoming exceptionally accurate in identifying people from images. However, this comes at the cost of privacy for users of online services with photo management (e.g. social media platforms). Particularly troubling is the ability to leverage unsupervised learning to recognize faces even when the user has not labeled their images. In this paper we propose Ulixes, a strategy to generate visually non-invasive facial noise masks that yield adversarial examples, preventing the formation of identifiable user clusters in the embedding space of facial encoders. This is applicable even when a user is unmasked and labeled images are available online. We demonstrate the effectiveness of Ulixes by showing that various classification and clustering methods cannot reliably label the adversarial examples we generate. We also study the effects of Ulixes in various black-box settings and compare it to the current state of the art in adversarial machine learning. Finally, we challenge the effectiveness of Ulixes against adversarially trained models and show that it is robust to countermeasures.

Telemedicine as technoinnovation to tackle COVID-19: A bibliometric analysis

Telemedicine has become fundamental for the challenges posed to healthcare. This set of instruments turns pivotal for facing one of the most relevant emergencies in human history: the COVID-19 pandemic. The multisectoral crisis led to a vigorously sustained adoption of innovations, including telemedicine technology. Telehealth was proven, in this context, to be a relevant tool to reduce healthcare costs, reduce not-needed hospitalizations, and improve the results in health care. Some barriers such as the costs of technologies, patient privacy and technical literacy have slowed down telemedicine adoption. Amidst the COVID-19 era, telemedicine calls for a managerial duty to change healthcare's organizational models. The present work aims to explore the growing literature to illuminate the relationships between telemedicine, innovations and healthcare in the COVID-19 framework. A bibliometric analysis of the existing literature based on 285 published works in 2019–2020 is put forward with the aim to detect the relevant literature, themes and approaches on telemedicine and COVID-19. Making use of community detection on the co-occurrence keywords network, we identify the “semantic cores” in the literature representing the relevant results on critical themes. The sorting implications are important for researchers and policymakers by mapping the existing literature and results in evidence-based analysis. We provide the key communities as the “semantic core” of the publications and results for the considered period. This allows for future research to be oriented towards perduring health policies that could lead to the adoption of telemedicine technologies in a post-pandemic scenario.

The Dark Side of Using Online Social Networks

Research on online social networks (OSNs) has focused overwhelmingly on their benefits and potential, with their negative effects overlooked. This study builds on the limited existing work on the so-called ‘dark side’ of using OSNs. The authors conducted a systematic review of selected databases and identified 46 negative effects of using OSNs from the users’ perspective, which is a rich spectrum of users’ negative experiences. This article then proposed nomenclature and taxonomy for the dark side of using OSNs by grouping these negative effects into six themes: cost of social exchange, cyberbullying, low performance, annoying content, privacy concerns and security threats. This study then conducted structured interviews with experts to confirm the sense-making and validity of the proposed taxonomy. This study discusses the confirmed taxonomy and outlines directions for future research.

Blockchain Technology in Agriculture: A Case Study of Blockchain Start-Up Companies

Agriculture is one of the areas where blockchain technology could bring a revolution by solving the existing problem of Agri-product fraud, its traceability, price manipulation, and lack of customer trust in the product. This paper aims to demonstrate the potential application of blockchain technology in the agriculture industry and how it could address the existing issues by surveying the existing paper and following case studies of the blockchain start-up companies. Blockchain technology shows a promising approach to fostering a safer, better, more sustainable, and dependable agri-foods system in the future. While the application of blockchain in agriculture is in the initial phase and faces various issues like cost of implementation, privacy, security scalability, performance, and infancy, it can bring a revolution in the agriculture industry.

The Transformation of Data Marketing: How an Ethical Lens on Consumer Data Collection Shapes the Future of Marketing

With a mass of customer data at our fingertips and the ability to use it to individualize promotion strategies, marketing communications, and product offerings, marketing activities are becoming more and more tailored to the individual customer. However, as concerns about online privacy and the handling of personal data take on an ever-increasing significance, marketers must increasingly evaluate and adapt their personalization and data collection methods. As a result, there is an increasing demand to take a critical look at the collection of data for personalization processes from an ethical perspective and to consider implications for further initiatives to maintain consumers’ trust. This research study utilizes a systematic literature review approach to investigate the current state of knowledge on the tradeoff between personalization and customer privacy by synthesizing and integrating extant knowledge. From the results of the present study’s search process, 20 articles were selected and analyzed for this review. Findings emphasize the importance of strengthening consumer relationships by increasing consumer trust, loyalty, confidence, and emotional attachment through specific organizational activities. The adaptation of marketing-related activities can thereby create a competitive advantage for data-collecting companies, as consumer backlash and privacy concerns decrease, and the willingness to disclose data increases. The current research contributes to the field of marketing by reviewing the issue of increasing personalization at the cost of customer privacy and explores how the resulting ethical considerations may affect the future of marketing practices. It thereby serves to help marketeers to implement effective strategies to ensure customer relationships and the resulting willingness to disclose personal data for personalization processes.

The Bisq decentralised exchange: on the privacy cost of participation

The Bisq Trade Protocol and the Bisq DAO (decentralised autonomous organisation) are core components of Bisq, a decentralised cryptocurrency exchange. The Bisq Trade Protocol systematises the peer-to-peer trading of Bitcoin for other currencies and the Bisq DAO decentralises the governance and finance functions of the entire exchange. However, by following the Bisq Trade Protocol and interacting with the Bisq DAO, participants necessarily publish data to the Bitcoin blockchain and broadcast additional data to the Bisq peer-to-peer network. We examine the privacy cost to participants in sharing this data. Specifically, we use novel address clustering heuristics to construct the one-to-many mappings from participants to addresses on the Bitcoin blockchain and augment the address clusters with data stored within the Bisq peer-to-peer network. We describe address clustering heuristics for both the Bisq Trade Protocol and the Bisq DAO. We show that the heuristics aggregate activity performed by each participant: trading, voting, transfers, etc. We identify instances where participants are operating under multiple aliases, some of which are real-world names. We identify the dominant transactors and their role in a two-sided market. We conclude with suggestions to better protect the privacy of participants in the future.

The underdeterrence, underperformance response to privacy, data protection laws

The digital economy has been defined in the economic literature as one with near zero marginal cost, unmonetized services but also an escalating data flow. After a careful review of the most recent economic papers, we offer an alternative theory on the cost of privacy and data protection regulations. We have observed that the characteristics of the regulation lead not only to the amplification of costs that have been traditionally assigned as variable costs by the literature, but also of costs that used to be fixed but have been outsourced in the digital economy, meaning that significant new variable costs might trigger diseconomies of scale. At the same time, privacy and data protection regulations have created incentives that are making the dominant firms insource, in what seems to be a way back to increased sunk fixed costs for these firms. Having all that in mind, we claim that the perception of deterrence and compliance costs has affected how firms might decide to incur higher risks to avoid costs. Although compliance costs are high, we claim that an efficient implementation of the regulation avoids much of these costs. Our claim is supported by evidence that a relevant share of the regulatory costs are now variable costs, leaving room for at least two efficient strategies that medium-sized firms might implement in order to avoid them. First, firms can lower the volumes of data that they use without significantly impairing the predictive functions of their algorithms. Second, firms can invest in security at a comparatively lower degree than dominant firms considering their lower exposure to strong regulatory action.

Enhanced privacy preserving and truth discovery method for 5G and beyond vehicle crowd sensing systems

As 5G networks are being deployed the key constraints of 5G networks have been highlighted, which encourages the probing research of 5G and beyond networks as the next-generation solutions. Especially for vehicular networks which require a network topology with very low latency and high reliability. 5G Vehicular networks offer higher reliability, lower latency, and secure and efficient communication services but these advancements come at the cost of new security and privacy concerns. Especially in the case of a high number of connected vehicles which leads to the concept of Vehicle Crowd Sensing System (VCS) in 5G enabled Vehicular Ad hoc networks (VANETS). Hence, due to this, the number of connected vehicle nodes is also soaring drastically which is directly proportional to the expansion of the Vehicle Crowd Sensing System (VCS). These vehicles are equipped with the latest gadgetries and a large number of sensors, which works simultaneously for processing at end-user utility applications e.g. navigation, predictions, and traffic monitoring. Therefore, in this scenario, all vehicle nodes upload their sole respective data to the cloud for computing which arise the need for reliability and privacy. Thus, Privacy-Preserving Truth Discovery (PPTD) is one such scheme that distills reliable information with privacy from raw signals acquired from different sources. Currently, PPTD brings substantial overhead in terms of consumed data, computation at the Vehicle level and requires each vehicle to remain online till the completion of uploading at the cloud server. So, in this paper, we explain a newly proposed novel Efficient Privacy-Preserving Truth Discovery (EPTD) entailing double masking for VCS and allowing Vehicles to go offline anytime during upload processing while ensuring privacy and accuracy at the same time. Simulation results have shown that the proposed work has outperformed all the previously existing methods.

ABCDP: Approximate Bayesian Computation with Differential Privacy.

We developed a novel approximate Bayesian computation (ABC) framework, ABCDP, which produces differentially private (DP) and approximate posterior samples. Our framework takes advantage of the sparse vector technique (SVT), widely studied in the differential privacy literature. SVT incurs the privacy cost only when a condition (whether a quantity of interest is above/below a threshold) is met. If the condition is sparsely met during the repeated queries, SVT can drastically reduce the cumulative privacy loss, unlike the usual case where every query incurs the privacy loss. In ABC, the quantity of interest is the distance between observed and simulated data, and only when the distance is below a threshold can we take the corresponding prior sample as a posterior sample. Hence, applying SVT to ABC is an organic way to transform an ABC algorithm to a privacy-preserving variant with minimal modification, but yields the posterior samples with a high privacy level. We theoretically analyzed the interplay between the noise added for privacy and the accuracy of the posterior samples. We apply ABCDP to several data simulators and show the efficacy of the proposed framework.

An Improved D-SMART Algorithm for Wireless Sensor Networks

In this paper, a secure and low energy dynamic slicing algorithm, namely, Improved D-SMART (IM-D-SMART) based on the Data Aggregation Protocol on Slice Mix Aggregate (D-SMART) is proposed to improve the security and confidentiality of wireless sensor networks and reduce the energy consumption of nodes in data collection and transmission in wireless sensor networks. According to the importance of data, the residual energy of nodes and the relative density of nodes, the data are dynamically partitioned to improve the D-SMART algorithm. Simultaneously, sending negative number splicing is used to compensate for the loss caused by the collision of data transmission between nodes. The simulation results show that IM-D-SMART outperforms D-SMART in terms of computation, privacy and communication cost.

Incentivizing Differentially Private Federated Learning: A Multidimensional Contract Approach

Federated learning is a promising tool in the Internet-of-Things (IoT) domain for training a machine learning model in a decentralized manner. Specifically, the data owners (e.g., IoT device consumers) keep their raw data and only share their local computation results to train the global model of the model owner (e.g., an IoT service provider). When executing the federated learning task, the data owners contribute their computation and communication resources. In this situation, the data owners have to face privacy issues where attackers may infer data property or recover the raw data based on the shared information. Considering these disadvantages, the data owners will be reluctant to use their data to participate in federated learning without a well-designed incentive mechanism. In this article, we deliberately design an incentive mechanism jointly considering the task expenditure and privacy issue of federated learning. Based on a differentially private federated learning (DPFL) framework that can prevent the privacy leakage of the data owners, we model the contribution as well as the computation, communication, and privacy costs of each data owner. The three types of costs are data owners' private information unknown to the model owner, which thus forms an information asymmetry. To maximize the utility of the model owner under such information asymmetry, we leverage a 3-D contract approach to design the incentive mechanism. The simulation results validate the effectiveness of the proposed incentive mechanism with the DPFL framework compared to other baseline mechanisms.

The Dark Side of Using Online Social Networks

Research on online social networks (OSNs) has focused overwhelmingly on their benefits and potential, with their negative effects overlooked. This study builds on the limited existing work on the so-called ‘dark side’ of using OSNs. The authors conducted a systematic review of selected databases and identified 46 negative effects of using OSNs from the users’ perspective, which is a rich spectrum of users’ negative experiences. This article then proposed nomenclature and taxonomy for the dark side of using OSNs by grouping these negative effects into six themes: cost of social exchange, cyberbullying, low performance, annoying content, privacy concerns and security threats. This study then conducted structured interviews with experts to confirm the sense-making and validity of the proposed taxonomy. This study discusses the confirmed taxonomy and outlines directions for future research.

Mobile cloud computing apps and information disclosure: the moderating roles of dispositional and behaviour-based traits

ABSTRACT The paradigm of developing mobile apps has shifted from native apps that store data on mobile devices to mobile cloud computing (MCC) apps that send data to the cloud. Transferring users’ data to the cloud provides several benefits, such as larger storage capacity and simultaneous access by multiple devices and users. However, storing data in the cloud also raises privacy concerns as users do not have direct control over their data. This study reports a privacy cost–benefit analysis including the moderating effects of dispositional traits (i.e. two personality meta-traits: stability and plasticity) and the behaviour-based trait (i.e. use experience) to understand information disclosure behaviour. The empirical study is based on a scenario-based survey (n = 807) from a diverse sample of MCC apps users. The results support the moderating effects of personality meta-traits; stability and plasticity differentially moderate the effects of perceived privacy risk and perceived value of data transfer to the cloud on information disclosure behaviour. Contradictory to prior research, prior use experience does not moderate the effects of the cost–benefit perceptions. The implications of research and practice are discussed.

A market-based rationale for the privacy paradox

Moving beyond the current focus on the individual as the unit of analysis in the privacy paradox, this article examines the misalignment between privacy attitudes and online behaviors at the level of society as a collective. I draw on Facebook’s market performance to show how despite concerns about privacy, market structures drive user, advertiser and investor behaviors to continue to reward corporate owners of social media platforms. In this market-oriented analysis, I introduce the metaphor of elasticity to capture the responsiveness of demand for social media to the data (price) charged by social media companies. Overall, this article positions social media as inelastic, relative to privacy costs; highlights the role of the social collective in the privacy crises; and ultimately underscores the need for structural interventions in addressing privacy risks.

Learning Model-Based Privacy Protection under Budget Constraints

Protecting privacy in gradient-based learning has become increasingly critical as more sensitive information is being used. Many existing solutions seek to protect the sensitive gradients by constraining the overall privacy cost within a constant budget, where the protection is hand-designed and empirically calibrated to boost the utility of the resulting model. However, it remains challenging to choose the proper protection adapted for specific constraints so that the utility is maximized. To this end, we propose a novel Learning-to-Protect algorithm that automatically learns a model-based protector from a set of non-private learning tasks. The learned protector can be applied to private learning tasks to improve utility within the specific privacy budget constraint. Our empirical studies on both synthetic and real datasets demonstrate that the proposed algorithm can achieve a superior utility with a given privacy constraint and generalize well to new private datasets distributed differently as compared to the hand-designed competitors.