Differentially Private SQL with Bounded User Contribution

Wilson, Royce J; Zhang, Celia Yuxin; Lam, William H. K.; Desfontaines, Damien; Simmons-Marengo, Daniel; Gipson, Bryant

doi:10.2478/popets-2020-0025

Cited by 84 publications

(82 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To produce this data set, machine learning was applied to log data to automatically segment it into semantic trips 48 . To provide strong privacy guarantees, all trips were anonymized and aggregated using a differentially private mechanism 49 to aggregate flows over time (https://policies.google.com/ technologies/anonymization). This research is done on the resulting heavily aggregated and differentially private data.…”

mentioning

confidence: 99%

Crowding and the shape of COVID-19 epidemics

Rader

Scarpino

Nande

et al. 2020

Nat Med

236

192

View full text Add to dashboard Cite

mentioning

confidence: 99%

Crowding and the shape of COVID-19 epidemics

Rader

Scarpino

Nande

et al. 2020

Nat Med

236

192

View full text Add to dashboard Cite

“…To produce this dataset, machine learning is applied to logs data to automatically segment it into semantic trips 42 . To provide strong privacy guarantees, all trips were anonymized and aggregated using a differentially private mechanism 43 to aggregate flows over time (see ref. 44 ).…”

Section: Methodsmentioning

confidence: 99%

“…We then remove all metrics for which the noisy number of users is lower than 100, following the process described in ref. 43 , and publish the rest. This yields that each metric we publish satisfies (ϵ,δ)-differential privacy with values defined above.…”

Section: Methodsmentioning

confidence: 99%

Forecasting influenza activity using machine-learned mobility map

et al. 2021

Self Cite

View full text Add to dashboard Cite

Human mobility is a primary driver of infectious disease spread. However, existing data is limited in availability, coverage, granularity, and timeliness. Data-driven forecasts of disease dynamics are crucial for decision-making by health officials and private citizens alike. In this work, we focus on a machine-learned anonymized mobility map (hereon referred to as AMM) aggregated over hundreds of millions of smartphones and evaluate its utility in forecasting epidemics. We factor AMM into a metapopulation model to retrospectively forecast influenza in the USA and Australia. We show that the AMM model performs on-par with those based on commuter surveys, which are sparsely available and expensive. We also compare it with gravity and radiation based models of mobility, and find that the radiation model’s performance is quite similar to AMM and commuter flows. Additionally, we demonstrate our model’s ability to predict disease spread even across state boundaries. Our work contributes towards developing timely infectious disease forecasting at a global scale using human mobility datasets expanding their applications in the area of infectious disease epidemiology.

show abstract

“…In [22], personal data is k-anonymized [23,24] before initiating the discovery process. The authors in [25] use a differential privacy agent [32][33][34][35] to provide noisy statistics on process traces to an untrusted process mining technique thereby preserving privacy of information contained in the event log. The query-synthesis mechanism and therefore the utility of the resulting discovered process model are improvable.…”

Section: Sn Computer Sciencementioning

confidence: 99%

On Enabling GDPR Compliance in Business Processes Through Data-Driven Solutions

Zaman

Hassani

2020

SN COMPUT. SCI.

View full text Add to dashboard Cite

The collection and long-term retention of excessive data enables organisations to process data for insights in non-primary processes. The discovery of insights is promoted to be useful both for organisations and the customers. However, long-term possession of data on one hand risks the privacy of data belonging beings in cases of data breaches and on the other hand results in the customers distrust. General Data Protection Regulation (GDPR) abstractly defined the data processing boundaries of the personal data of European Union's citizens. The processing principles of GDPR, in line with the spirit of privacy by design and default, provide directions on the collection, storage, and processing of personal data. Concomitantly, the data subject rights provide customers with necessary control over their personal data stationed at the data controller's premises. The accountability principle of GDPR requires compliance in place and also the ability to demonstrate it. In this work, we are providing three solutions to enable GDPR compliance in business processes. First, we are proposing intra-process data degradation, a solution for continuous data minimisation during the course of business processes. The proposed approach results in reduced data maintenance and breach losses. Second, we adapt process mining techniques for ascertaining compliance of business process execution to data subject rights. Finally, we present a scheme to utilise differential privacy technique to enable GDPR-compliant business process discovery. Additionally, we offer links to two effective tools that demonstrate our first and second contributions.

show abstract

Differentially Private SQL with Bounded User Contribution

Cited by 84 publications

References 32 publications

Crowding and the shape of COVID-19 epidemics

Crowding and the shape of COVID-19 epidemics

Forecasting influenza activity using machine-learned mobility map

On Enabling GDPR Compliance in Business Processes Through Data-Driven Solutions

Contact Info

Product

Resources

About