Collecting and Analyzing Data from Smart Device Users with Local Differential Privacy

Abstract

Organizations with a large user base, such as Samsung and Google, can potentially benefit from collecting and mining users' data. However, doing so raises privacy concerns, and risks accidental privacy breaches with serious consequences. Local differential privacy (LDP) techniques address this problem by only collecting randomized answers from each user, with guarantees of plausible deniability; meanwhile, the aggregator can still build accurate models and predictors by analyzing large amounts of such randomized data. So far, existing LDP solutions either have severely restricted functionality, or focus mainly on theoretical aspects such as asymptotical bounds rather than practical usability and performance. Motivated by this, we propose Harmony, a practical, accurate and efficient system for collecting and analyzing data from smart device users, while satisfying LDP. Harmony applies to multi-dimensional data containing both numerical and categorical attributes, and supports both basic statistics (e.g., mean and frequency estimates), and complex machine learning tasks (e.g., linear regression, logistic regression and SVM classification). Experiments using real data confirm Harmony's effectiveness.

arXiv comment11 pages

References in corpus (2)

• Crowd-ML: A Privacy-Preserving Learning Framework for a Crowd of Smart Devices
• Differentially Private Multi-party Computation: Optimality of Non-Interactive Randomized Response

Cited by in corpus (30)

• A Comprehensive Survey on Local Differential Privacy Toward Data Statistics and Analysis
• LDP-IDS: Local Differential Privacy for Infinite Data Streams
• Local Differential Privacy and Its Applications: A Comprehensive Survey
• Breaking the Communication-Privacy-Accuracy Trilemma
• Local Differential Privacy: a tutorial
• LDP-FL: Practical Private Aggregation in Federated Learning with Local Differential Privacy
• Improving the utility of locally differentially private protocols for longitudinal and multidimensional frequency estimates
• Successive Point-of-Interest Recommendation with Local Differential Privacy
• Random Sampling Plus Fake Data: Multidimensional Frequency Estimates With Local Differential Privacy
• On the Power of Multiple Anonymous Messages
• Locally Differentially Private Heavy Hitter Identification
• PCKV: Locally Differentially Private Correlated Key-Value Data Collection with Optimized Utility
• Conditional Analysis for Key-Value Data with Local Differential Privacy
• Residue-based Label Protection Mechanisms in Vertical Logistic Regression
• Locally Differentially Private Naive Bayes Classification
• Chasing Your Long Tails: Differentially Private Prediction in Health Care Settings
• FedSel: Federated SGD under Local Differential Privacy with Top-k Dimension Selection
• Locally Differentially Private Data Collection and Analysis
• Optimizing Locally Differentially Private Protocols
• Aggregating Votes with Local Differential Privacy: Usefulness, Soundness vs. Indistinguishability
• Privacy-Preserving Image Classification in the Local Setting
• Revealing the True Cost of Locally Differentially Private Protocols: An Auditing Perspective
• Federated learning with differential privacy and an untrusted aggregator
• Decentralized Differentially Private Without-Replacement Stochastic Gradient Descent
• BiSample: Bidirectional Sampling for Handling Missing Data with Local Differential Privacy
• AHEAD: Adaptive Hierarchical Decomposition for Range Query under Local Differential Privacy
• Locally Differentially Private Bayesian Inference
• Marginal Release Under Local Differential Privacy
• Privacy Enhancement via Dummy Points in the Shuffle Model
• Selective MPC: Distributed Computation of Differentially Private Key-Value Statistics