Secure Bayesian Federated Analytics for Privacy-Preserving Trend Detection

TL;DR

This paper introduces SAFE, a privacy-preserving Bayesian federated analytics protocol for trend detection that securely aggregates local data to identify trending keywords while maintaining user privacy.

Contribution

It presents a novel secure Bayesian approach and protocol for federated trend detection, reducing computational load and enhancing privacy for practical deployment.

Findings

SAFE achieves privacy-preserving trend detection in federated settings

The protocol reduces computational burden on users and aggregator

Experimental results demonstrate effective trend detection with privacy guarantees

Abstract

Federated analytics has many applications in edge computing, its use can lead to better decision making for service provision, product development, and user experience. We propose a Bayesian approach to trend detection in which the probability of a keyword being trendy, given a dataset, is computed via Bayes' Theorem; the probability of a dataset, given that a keyword is trendy, is computed through secure aggregation of such conditional probabilities over local datasets of users. We propose a protocol, named SAFE, for Bayesian federated analytics that offers sufficient privacy for production grade use cases and reduces the computational burden of users and an aggregator. We illustrate this approach with a trend detection experiment and discuss how this approach could be extended further to make it production-ready.