Privacy Regulations Research Articles

Fractals as Pre-Training Datasets for Anomaly Detection and Localization

Anomaly detection is crucial in large-scale industrial manufacturing as it helps to detect and localize defective parts. Pre-training feature extractors on large-scale datasets is a popular approach for this task. Stringent data security, privacy regulations, high costs, and long acquisition time hinder the development of large-scale datasets for training and benchmarking. Despite recent work focusing primarily on the development of new anomaly detection methods based on such extractors, not much attention has been paid to the importance of the data used for pre-training. This study compares representative models pre-trained with fractal images against those pre-trained with ImageNet, without subsequent task-specific fine-tuning. We evaluated the performance of eleven state-of-the-art methods on MVTecAD, MVTec LOCO AD, and VisA, well-known benchmark datasets inspired by real-world industrial inspection scenarios. Further, we propose a novel method to create a dataset by combining the dynamically generated fractal images creating a “Multi-Formula” dataset. Even though pre-training with ImageNet leads to better results, fractals can achieve close performance to ImageNet under proper parametrization. This opens up the possibility for a new research direction where feature extractors could be trained on synthetically generated abstract datasets mitigating the ever-increasing demand for data in machine learning while circumventing privacy and security concerns.

Data Privacy Concerns and their Impact on Consumer Trust in Digital Marketing

The rise of digital marketing has transformed consumer-brand interactions through personalized strategies and targeted ads driven by extensive consumer data. However, increased data collection has sparked privacy concerns, affecting consumer trust. This study examines how dataprivacy concerns impact consumer trust, engagement, and willingness to share information in digital marketing. Key privacy issues include over-collection, lack of transparency, unauthorized sharing, and data breaches. Using a mixed-methods approach, with quantitative surveys of 600 participants and qualitative interviews with 30 individuals, the study measures privacy concerns, trust, and engagement levels. Findings reveal a strong negative correlation between privacy concerns and trust, with transparency in data practices and regulatory compliance (GDPR,CCPA) identified as critical trust factors. Brands that prioritize transparency, data control, and compliance foster greater consumer trust, addressing the privacy- personalization paradox and aligning with consumer expectations for ethical data use. This study offers insights into privacy-trust dynamics and provides recommendations for privacy-centered marketing strategies. Keywords:Data Privacy ,Consumer Trust, User Consent ,Data Transparency, Data Collection Practices, Third-Party Data Sharing ,Data Breaches, Privacy Regulations (e.g., GDPR, CCPA).

Integrating Blockchain and Homomorphic Encryption to Enhance Security and Privacy in Project Management and Combat Counterfeit Goods in Global Supply Chain Operations

The proliferation of counterfeit goods and data privacy concerns in global supply chain operations poses a significant risk to organizations and end-users alike. This paper explores the integration of blockchain technology and homomorphic encryption as dual mechanisms to enhance security and privacy in project management and supply chain operations. Blockchain offers an immutable, decentralized ledger that provides traceability and authenticity across supply chain nodes, while homomorphic encryption enables secure data processing and analysis without compromising sensitive information. The synergy of these technologies presents a robust framework to address counterfeit risks, facilitating end-to-end product authentication and allowing for secure, real-time collaboration in project management settings. This integration minimizes vulnerabilities in data handling and improves compliance with privacy regulations by securing data at rest and in transit. By evaluating case studies and recent advancements, this paper highlights the potential of blockchain and homomorphic encryption to redefine security protocols, fostering a trustworthy and resilient supply chain ecosystem. Future research directions are recommended to expand the practical applications and address scalability challenges associated with these emerging technologies.

Recent Advancements in Federated Learning: State of the Art, Fundamentals, Principles, IoT Applications and Future Trends

Federated learning (FL) is creating a paradigm shift in machine learning by directing the focus of model training to where the data actually exist. Instead of drawing all data into a central location, which raises concerns about privacy, costs, and delays, FL allows learning to take place directly on the device, keeping the data safe and minimizing the need for transfer. This approach is especially important in areas like healthcare, where protecting patient privacy is critical, and in industrial IoT settings, where moving large numbers of data is not practical. What makes FL even more compelling is its ability to reduce the bias that can occur when all data are centralized, leading to fairer and more inclusive machine learning outcomes. However, it is not without its challenges—particularly with regard to keeping the models secure from attacks. Nonetheless, the potential benefits are clear: FL can lower the costs associated with data storage and processing, while also helping organizations to meet strict privacy regulations like GDPR. As edge computing continues to grow, FL’s decentralized approach could play a key role in shaping how we handle data in the future, moving toward a more privacy-conscious world. This study identifies ongoing challenges in ensuring model security against adversarial attacks, pointing to the need for further research in this area.

Federated learning with tensor networks: a quantum AI framework for healthcare

Abstract The healthcare industries frequently handle sensitive and proprietary data, and due to strict
privacy regulations, they are often reluctant to share data directly. In today’s context, Federated
Learning (FL) stands out as a crucial remedy, facilitating the rapid advancement of distributed machine
learning while effectively managing critical concerns regarding data privacy and governance.
The fusion of federated learning and quantum computing represents a groundbreaking interdisciplinary
approach with immense potential to revolutionize various industries, from healthcare to
finance. In this work, we propose a federated learning framework based on quantum tensor networks
that takes advantage of the principles of many-body quantum physics. Currently, there are
no known classical tensor networks implemented in federated settings. Furthermore, we investigated
the effectiveness and feasibility of the proposed framework by conducting a differential privacy analysis
to ensure the security of sensitive data across healthcare institutions. Experiments on popular
medical image datasets show that the federated quantum tensor network model achieved a mean
receiver-operator characteristic area under the curve (ROC-AUC) of 91-98%, outperforming several
state-of-the-art federated learning methods while using fewer parameters. Experimental results
demonstrate that the quantum federated global model, consisting of highly entangled tensor network
structures, showed better generalization and robustness and achieved higher testing accuracy,
surpassing the performance of locally trained clients under unbalanced data distributions among
healthcare institutions.

Multidimensional Impact Analysis of Public Policies on the European Economy: Taking EU Member States as an Example

Since modern times, with the close trade relations between European countries, they have begun the process of integration in order to strengthen economic cooperation and political unity. In the process of integration, problems such as economic inequality, political differences, and external interference continue to emerge. In the process of solving these problems, public policy plays a crucial role. The European Union promotes cooperation and coordination among member states by formulating and implementing a series of public policies, and promotes the deep development of European integration. This enables European countries to smoothly navigate through various crises. This article mainly discusses the impact of European public policies on the Eurozone economy, including problems in the European economic governance system, exploration and analysis of fiscal integration implementation, policies proposed in response to the rigidity of the Italian labor market, the impact of EU data privacy and financial market regulation policies, and future prospects.

Evaluating the Role of Data Privacy Regulations in Secure Software Development Life Cycles (SDLC)

Evaluating the Role of Data Privacy Regulations in Secure Software Development Life Cycles (SDLC)

INTERNATIONAL COMMUNITY IN THE GLOBAL DIGITAL ECONOMY: A CASE STUDY ON THE AFRICAN DIGITAL TRADE FRAMEWORK

Abstract Global digital integration is desirable and perhaps even inevitable for most States. However, there is currently no systematic framework or narrative to drive such integration in trade agreements. This article evaluates whether community values can offer a normative foundation for rules governing digital trade. It uses the African Continental Free Trade Area (AfCFTA) Digital Trade Protocol as a case study and argues that identifying and solidifying the collective needs of the African region through this instrument will be key to shaping an inclusive and holistic regional framework. These arguments are substantiated by analysis of the regulation of cross-border data flows, privacy and cybersecurity.

Leveraging Federated Learning for Privacy-Preserving Analysis of Multi-Institutional Electronic Health Records in Rare Disease Research

This research announces that the fresh federated learning structure is designed to enhance the privacy-preserving analysis of electronic health records (EHRs), and multiple institutions in this framework permit secure collaboration among institutions, allowing them to train machine-learning replicas without directly sharing patient data. We implemented and evaluated numerous machine-learning models to forecast patient treatment needs, including Logistic Regression, Decision-Tree-Classifiers, Support-Vectors-Classifiers, Random-Forests, and Stacking-Classifiers. The Random Forest classifier achieved the best performance with an accuracy of 90% and an F1 score of 80%, demonstrating that it handled complex and imbalanced datasets. This FL-based approach not only complies with privacy regulations such as HIPAA and GDPR but also overcomes significant challenges in data sharing, making it ideal for rare disease research. By enabling secure data aggregation across institutions, the framework significantly enhances the ability to study rare diseases and accelerates the discovery of new treatments. Future directions include extending this framework to other areas of healthcare and incorporating advanced machine-learning techniques to enhance its capabilities further. This research sets the new standard for secure and collaborative healthcare data analysis and promotes innovation and ethical practices in rare disease research.

Synthetic data generation with hybrid quantum-classical models for the financial sector

Abstract Data integrity and privacy are critical concerns in the financial sector. Traditional methods of data collection face challenges due to privacy regulations and time-consuming anonymization processes. In collaboration with Banco BV, we trained a hybrid quantum-classical generative adversarial network (HQGAN), where a quantum circuit serves as the generator and a classical neural network acts as the discriminator, to generate synthetic financial data efficiently and securely. We compared our proposed HQGAN model with a fully classical GAN by evaluating loss convergence and the MSE distance between the synthetic and real data. Although initially promising, our evaluation revealed that HQGAN failed to achieve the necessary accuracy to understand the intricate patterns in financial data. This outcome underscores the current limitations of quantum-inspired methods in handling the complexities of financial datasets. Graphical abstract

Clinical information extraction for lower-resource languages and domains with few-shot learning using pretrained language models and prompting

Abstract A vast amount of clinical data are still stored in unstructured text. Automatic extraction of medical information from these data poses several challenges: high costs of clinical expertise, restricted computational resources, strict privacy regulations, and limited interpretability of model predictions. Recent domain adaptation and prompting methods using lightweight masked language models showed promising results with minimal training data and allow for application of well-established interpretability methods. We are first to present a systematic evaluation of advanced domain-adaptation and prompting methods in a lower-resource medical domain task, performing multi-class section classification on German doctor’s letters. We evaluate a variety of models, model sizes (further-pre)training and task settings, and conduct extensive class-wise evaluations supported by Shapley values to validate the quality of small-scale training data and to ensure interpretability of model predictions. We show that in few-shot learning scenarios, a lightweight, domain-adapted pretrained language model, prompted with just 20 shots per section class, outperforms a traditional classification model, by increasing accuracy from $48.6\%$ to $79.1\%$ . By using Shapley values for model selection and training data optimization, we could further increase accuracy up to $84.3\%$ . Our analyses reveal that pretraining of masked language models on general-language data is important to support successful domain-transfer to medical language, so that further-pretraining of general-language models on domain-specific documents can outperform models pretrained on domain-specific data only. Our evaluations show that applying prompting based on general-language pretrained masked language models combined with further-pretraining on medical-domain data achieves significant improvements in accuracy beyond traditional models with minimal training data. Further performance improvements and interpretability of results can be achieved, using interpretability methods such as Shapley values. Our findings highlight the feasibility of deploying powerful machine learning methods in clinical settings and can serve as a process-oriented guideline for lower-resource languages and domains such as clinical information extraction projects.

How does the consent model in Australia’s Privacy Act 1988 (Cth) undermine our human right to privacy?

ABSTRACT Cambridge Analytica was a political consulting business that became known for its use of personal data obtained from social media platform Facebook, to drive election campaigns on behalf of its clients. Their most famous client was the winner of the 2016 US Presidential election. When it was realised that Cambridge Analytica had acquired Australian Facebook users’ data, the Australian privacy regulator started litigation against Facebook to enforce the Privacy Act 1988 (Cth). This article examines how the Facebook litigation and subsequent law reform activities shed light on the central, yet fictitious role of the consent model. A historical and cultural analysis reveals how consent to surrender private information is a symptom of social conditions in which we take it for granted that individuals, rather than business, will bear the costs of poor privacy outcomes. Our collective ‘consent realism’ not only constrains the imagination of policy makers but interferes with a person’s autonomy in forming and developing opinions. However, a right to privacy can be more viably re-asserted, at least in response to the Cambridge Analytica incident, within the framework of Australia’s implied freedom of political communication. Consent realism can thus potentially be overcome in favour of human rights principles.

Horti-IoT: a new data framework for Dutch horticulture

Abstract Consistency issues limit the sharing of horticultural data across multiple systems, resulting in challenges for users to analyze data effectively across various systems utilizing artificial intelligence technology. Introducing data governance principles can help standardize and unify data practices, making it easier for analysts to locate, comprehend, transfer and integrate data from diverse sources to enable data-driven horticulture. Implementing data governance and principles specific to horticulture can assist in standardizing the layout and format of data structures from different sources. This study aims to propose a new governance framework, Horti-IoT, based on the Data Management Body of Knowledge and several structured frameworks for the Internet of Things (IoT) governance that will lead to data-driven horticulture. This study is empirical in nature. The Dutch horticulture stakeholders are involved in this initiative, providing the data, knowledge, and experiences needed for this study. The data stream from various sources, including camera images, sap flow sensors, climate sensors and manually measured growth data. The key findings following the implementation of the Horti-IoT framework’s principles are reduced workload for data analysts, efficiency in plant monitoring, savings time in pre-processing, enhanced water resource management, reduced system administrator contacts and compliance with General Data Privacy Regulation. The new proposed Horti-IoT framework, compatible with Dutch horticulture, is presented. The data were obtained from the Lab greenhouse at the World Horti Centre in the Netherlands, in the framework of the Regionale SIA RAAK MKB call March 2022-September 2024 subsidy funds for project title ‘Gewasgroei Goed Gemeten (GeGoGe). This project is a collaboration between three educational institutions. Inholland University of Applied Science, the Hague University of Applied Science, Lentiz Vocational School, and stakeholders.

A Review on Robust Credit Card Fraud Detection System Leveraging Big Data and Machine Learning

This review offers a detailed strategy to address the growing threat of credit card fraud in today's digital landscape. By utilizing Big Data analytics alongside machine learning methods, the system aims to transform fraud detection processes. It tackles the challenges arising from the increasing volume and complexity of credit card transactions, enabling the real-time detection and prevention of fraudulent actions. The system employs sophisticated machine learning algorithms to identify patterns and anomalies linked to fraudulent activities, allowing for proactive responses to emerging fraud tactics. Additionally, the system is optimized to handle and analyze large datasets efficiently, ensuring timely and precise detection of fraud. It also incorporates strong security protocols to protect sensitive customer data while adhering to privacy regulations. This review ultimately seeks to enhance the safety and reliability of electronic payments, protecting financial institutions and consumers from the harmful effects of credit card fraud.

Speaking in Private: Privacy Expectations Depend on Communication Modality

Consumers disclose personal information when they interact with connected technologies. The advent of voice technology has enabled consumers to interact with connected technologies not only through typing but also through speaking. The present research investigates whether consumers expect different levels of privacy for information they disclose via different communication modalities. The results of three studies suggest that consumers have more restrictive privacy expectations for information disclosed via speech as compared with text. The studies probe the viability of several mechanisms that may drive this effect and test practically relevant moderators. The results suggest that the effect is driven, at least in part, by increased feelings of ownership over content disclosed via speech as compared with text. Of relevance to multiple stakeholders, the article discusses implications for privacy regulation, privacy-preserving application design, targeted advertising, and the “privacy paradox.” This paper has been This paper was accepted by Catherine Tucker for the Special Issue on the Human-Algorithm Connection. Funding: This work was supported by Marketing Science Institute [4001309], NYU Stern Center for Global Economy and Business. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.03954 .

Efficient Credit Card Fraud Detection System Using Big Data and Machine Learning

This review offers a detailed strategy to address the growing threat of credit card fraud in today's digital landscape. By utilizing Big Data analytics alongside machine learning methods, the system aims to transform fraud detection processes. It tackles the challenges arising from the increasing volume and complexity of credit card transactions, enabling the real-time detection and prevention of fraudulent actions. The system employs sophisticated machine learning algorithms to identify patterns and anomalies linked to fraudulent activities, allowing for proactive responses to emerging fraud tactics. Additionally, the system is optimized to handle and analyze large datasets efficiently, ensuring timely and precise detection of fraud. It also incorporates strong security protocols to protect sensitive customer data while adhering to privacy regulations. This review ultimately seeks to enhance the safety and reliability of electronic payments, protecting financial institutions and consumers from the harmful effects of credit card fraud.

The potential of federated learning for self-configuring medical object detection in heterogeneous data distributions

Medical Object Detection (MOD) is a clinically relevant image processing method that locates structures of interest in radiological image data at object-level using bounding boxes. High-performing MOD models necessitate large datasets accurately reflecting the feature distribution of the corresponding problem domain. However, strict privacy regulations protecting patient data often hinder data consolidation, negatively affecting the performance and generalization of MOD models. Federated Learning (FL) offers a solution by enabling model training while the data remain at its original source institution. While existing FL solutions for medical image classification and segmentation demonstrate promising performance, FL for MOD remains largely unexplored. Motivated by this lack of technical solutions, we present an open-source, self-configuring and task-agnostic federated MOD framework. It integrates the FL framework Flower with nnDetection, a state-of-the-art MOD framework and provides several FL aggregation strategies. Furthermore, we evaluate model performance by creating simulated Independent Identically Distributed (IID) and non-IID scenarios, utilizing the publicly available datasets. Additionally, a detailed analysis of the distributions and characteristics of these datasets offers insights into how they can impact performance. Our framework’s implementation demonstrates the feasibility of federated self-configuring MOD in non-IID scenarios and facilitates the development of MOD models trained on large distributed databases.

Lobbying global venues: Sitting in or speaking out?

AbstractUnderstanding interest groups' participation in global policy processes is critical not least because of an increasing shift in policy‐making powers to global institutions. This paper contributes to existing research by examining advocacy efforts at the global level and proposing a novel argument linking the degree of policy complexity and the amount of groups' resources to lobbying strategies. Specifically, it argues that interest groups invest in both inside (“sitting in”) and outside (“speaking up”) lobbying strategies when the policy at stake is complex and they have more resources. This theory is tested using extensive and novel data spanning interest groups' lobbying efforts on global Internet privacy regulation.

Conversation-Related Advertising and Electronic Eavesdropping: Mapping Perceptions of Phones Listening for Advertising in the United States, the Netherlands, and Poland

People report receiving ads on their mobile device that are seemingly related to previous offline conversations (i.e., conversation-related advertising). They may think that this is because their electronic devices are eavesdropping (i.e., e-eavesdropping). To gain insights into the scope and characteristics of conversation-related advertising and e-eavesdropping beliefs, we conducted a survey in the United States ( n = 300), the Netherlands ( n = 293), and Poland ( n = 293). These countries were chosen based on their differences in privacy regulations and history with state surveillance. We find that belief in conversation-related advertising is a widespread cross-country phenomenon, which is higher in the United States compared with European countries. In addition, between half and two-thirds of respondents believe e-eavesdropping is a likely explanation for it. We find that social media is a main contributing factor through which people hear about and most often see conversation-related advertising. Moreover, in response to such advertising, respondents experience more negative than positive affect. The results show that this is a prevalent and timely phenomenon that warrants more research. This carries implications for the (social) media industry and regulators, as it highlights the importance of transparency and (insufficient) individual understanding of data collection and processing.

Johnny Still Can't Opt-out: Assessing the IAB CCPA Compliance Framework

The privacy laws and regulations that govern the collection, sharing, and selling of online data are changing. In the U.S., California adopted the California Consumer Privacy Act (CCPA) and the California Privacy Rights Act (CPRA), and twelve other U.S. states have adopted similar laws. Industry has responded by developing technical standards for collecting and disseminating consent information, such as the IAB CCPA Compliance Framework. While publishers are adopting this framework and the IAB is extending it to cover privacy laws in other U.S. states, recent work has observed that opt-out signals are not being honored under the framework. In this study, we take a deep dive into the IAB CCPA Compliance Framework to measure end-to-end flows of consent information and better understand why opt-out signals are not being honored. Using data crawled from top websites under different experimental conditions, we examine overall adoption of the framework, the flow of consent information from publishers to third parties and between third parties, and finally the reach of opt-out signals. Our results uncover numerous issues with the adoption and implementation of the framework that prevent users' consent choices from being honored by third parties.