PANOLA: A Personal Assistant for Supporting Users in Preserving Privacy

Privacy is the right of individuals to keep personal information to themselves. When individuals use online systems, they should be given the right to decide what information they would like to share and what to keep private. When a piece of information pertains only to a single individual, preserving privacy is possible by providing the right access options to the user. However, when a piece of information pertains to multiple individuals, such as a picture of a group of friends or a collaboratively edited document, deciding how to share this information and with whom is challenging. The problem becomes more difficult when the individuals who are affected by the information have different, possibly conflicting privacy constraints. Resolving this problem requires a mechanism that takes into account the relevant individuals’ concerns to decide on the privacy configuration of information. Because these decisions need to be made frequently (i.e., per each piece of shared content), the mechanism should be automated. This article presents a personal assistant to help end-users with managing the privacy of their content. When some content that belongs to multiple users is about to be shared, the personal assistants of the users employ an auction-based privacy mechanism to regulate the privacy of the content. To do so, each personal assistant learns the preferences of its user over time and produces bids accordingly. Our proposed personal assistant is capable of assisting users with different personas and thus ensures that people benefit from it as they need it. Our evaluations over multiagent simulations with online social network content show that our proposed personal assistant enables privacy-respecting content sharing.

The experiences of 33 national COVID-19 dashboard teams during the first year of the pandemic in the WHO European Region: a qualitative study

Background: Governments across the WHO European Region prioritized dashboards for reporting COVID-19 data. The ubiquitous use of dashboards for public reporting is novel. This study explores the development of COVID-19 dashboards during the pandemic's first year and common barriers, enablers and lessons from the experiences of teams responsible for their development. Methods: Multiple methods were applied to identify and recruit COVID-19 dashboard teams using a purposive, quota sampling approach. Semi-structured group interviews were conducted between April-June 2021. Using elaborative coding and thematic analysis, descriptive and explanatory themes were derived from interview data. A validation workshop with study participants was held in June 2021. Results: Eighty informants, representing 33 national COVID-19 dashboard teams across the WHO European Region participated. Most dashboards were launched swiftly in the first months of the pandemic, between February-May 2020. The urgency, intense workload, limited human resources, data and privacy constraints, and public scrutiny were common to the initial development stage. Themes related to barriers or enablers were identified pertaining to the pre-pandemic context, pandemic itself, people and processes, software, data, and users. Lessons emerged around the themes of simplicity, trust, partnership, software and data, and change. Conclusions: COVID-19 dashboards were developed in a learning-by-doing approach. The experiences of teams signal initial under-preparedness was compensated by high-level political endorsement, the professionalism of teams, accelerated data improvements, and immediate support of commercial software solutions. To leverage the full potential of dashboards, investments are needed at team-, national- and pan-European-level.

Gendered health, economic, social and safety impact of COVID-19 on adolescents and young adults in Nairobi, Kenya

Background Infectious disease outbreaks like COVID-19 and their mitigation measures can exacerbate underlying gender disparities, particularly among adolescents and young adults in densely populated urban settings. Methods An existing cohort of youth ages 16–26 in Nairobi, Kenya completed a phone-based survey in August-October 2020 (n = 1217), supplemented by virtual focus group discussions and interviews with youth and stakeholders, to examine economic, health, social, and safety experiences during COVID-19, and gender disparities therein. Results COVID-19 risk perception was high with a gender differential favoring young women (95.5% vs. 84.2%; p<0.001); youth described mixed concern and challenges to prevention. During COVID-19, gender symmetry was observed in constrained access to contraception among contraceptive users (40.4% men; 34.6% women) and depressive symptoms (21.8% men; 24.3% women). Gender disparities rendered young women disproportionately unable to meet basic economic needs (adjusted odds ratio [aOR] = 1.21; p<0.05) and in need of healthcare during the pandemic (aOR = 1.59; p<0.001). At a bivariate level, women had lower full decisional control to leave the house (40.0% vs. 53.2%) and less consistent access to safe, private internet (26.1% vs. 40.2%), while men disproportionately experienced police interactions (60.1%, 55.2% of which included extortion). Gender-specific concerns for women included menstrual hygiene access challenges (52.0%), increased reliance on transactional partnerships, and gender-based violence, with 17.3% reporting past-year partner violence and 3.0% non-partner sexual violence. Qualitative results contextualize the mental health impact of economic disruption and isolation, and, among young women, privacy constraints. Implications Youth and young adults face gendered impacts of COVID-19, reflecting both underlying disparities and the pandemic’s economic and social shock. Economic, health and technology-based supports must ensure equitable access for young women. Gender-responsive recovery efforts are necessary and must address the unique needs of youth.

Secure Deep Graph Generation with Link Differential Privacy

Many data mining and analytical tasks rely on the abstraction of networks (graphs) to summarize relational structures among individuals (nodes). Since relational data are often sensitive, we aim to seek effective approaches to generate utility-preserved yet privacy-protected structured data. In this paper, we leverage the differential privacy (DP) framework to formulate and enforce rigorous privacy constraints on deep graph generation models, with a focus on edge-DP to guarantee individual link privacy. In particular, we enforce edge-DP by injecting designated noise to the gradients of a link reconstruction based graph generation model, while ensuring data utility by improving structure learning with structure-oriented graph discrimination. Extensive experiments on two real-world network datasets show that our proposed DPGGAN model is able to generate graphs with effectively preserved global structure and rigorously protected individual link privacy.

Data privacy preservation algorithm with k-anonymity

With growing concern of data privacy violations, privacy preservation processes become more intense. The k-anonymity method, a widely applied technique, transforms the data such that the publishing datasets must have at least k tuples to have the same link-able attribute, quasi-identifiers, values. From the observations, we found that, in a certain domain, all quasi-identifiers of the datasets, can have the same data type. This type of attribute is considered as an Identical Generalization Hierarchy (IGH) data. An IGH data has a particular set of characteristics that could utilize for enhancing the efficiency of heuristic privacy preservation algorithms. In this paper, we propose a data privacy preservation heuristic algorithm on IGH data. The algorithm is developed from the observations on the anonymous property of the problem structure that can eliminate the privacy constraints consideration. The experiment results are presented that the proposed algorithm could effectively preserve data privacy and also reduce the number of visited nodes for ensuring the privacy protection, which is the most time-consuming process, compared to the most efficient existing algorithm by at most 21%.

Smart Parking with Fair Selection and Imposing Higher Privacy Constraints in Parking Owner and Driver Information

The use of private cars has enhanced the comfort of travel of individuals, but has proven to be a challenge for parking in congested downtown areas and metropolitans. This hike in the vehicle count has led to difficulty among the drivers to find a parking spot, exploiting resources and time. On the other hand, there are many idle private parking spots that remain inaccessible because of multiple reasons like unavailable owners, different open timings and so on. In order to prevent parking issues as well as to enable the use of private parking spots, smart parking applications that are easy to use by the drivers will prove to be highly effective. However, most parking lot owners and drivers face the threat of privacy which affects their willingness to participate while many others are located in a centralized location where the presence of malicious users is in plenty. In this proposed work, we have introduced a smart-parking system that is based on blockchain exhibiting qualities of privacy protection, reliability and fairness. To protect the privacy of users, vector-based encryption, bloom filters and group signatures are also insisted. This has helped us establish a more reliable smart parking system coupled with fair operation for smart contact. Experimental analysis of the real-world dataset indicates that the proposed work operates with high efficiency, establishing privacy protection, reliability and fairness.

Scalable Privacy-Preserving Distributed Learning

In this paper, we address the problem of privacy-preserving distributed learning and the evaluation of machine-learning models by analyzing it in the widespread MapReduce abstraction that we extend with privacy constraints. We design spindle (Scalable Privacy-preservINg Distributed LEarning), the first distributed and privacy-preserving system that covers the complete ML workflow by enabling the execution of a cooperative gradient-descent and the evaluation of the obtained model and by preserving data and model confidentiality in a passive-adversary model with up to N −1 colluding parties. spindle uses multiparty homomorphic encryption to execute parallel high-depth computations on encrypted data without significant overhead. We instantiate spindle for the training and evaluation of generalized linear models on distributed datasets and show that it is able to accurately (on par with non-secure centrally-trained models) and efficiently (due to a multi-level parallelization of the computations) train models that require a high number of iterations on large input data with thousands of features, distributed among hundreds of data providers. For instance, it trains a logistic-regression model on a dataset of one million samples with 32 features distributed among 160 data providers in less than three minutes.