Google COVID-19 搜索趋势症状数据集：匿名化过程描述（1.0 版）
密码学与安全
2020-09-04 v1
摘要
本报告描述了应用于 COVID-19 搜索趋势症状数据集初始版本（于 2020 年 9 月 2 日发布于 https://goo.gle/covid19symptomdataset）的聚合与匿名化过程，该公开数据集展示了 Google 症状（及部分相关主题）搜索的聚合、匿名化趋势。匿名化过程旨在以 ε = 1.68 的 ε-差分隐私保护每位用户的每日症状搜索活动。
引用
@article{arxiv.2009.01265,
title = {Google COVID-19 Search Trends Symptoms Dataset: Anonymization Process Description (version 1.0)},
author = {Shailesh Bavadekar and Andrew Dai and John Davis and Damien Desfontaines and Ilya Eckstein and Katie Everett and Alex Fabrikant and Gerardo Flores and Evgeniy Gabrilovich and Krishna Gadepalli and Shane Glass and Rayman Huang and Chaitanya Kamath and Dennis Kraft and Akim Kumok and Hinali Marfatia and Yael Mayer and Benjamin Miller and Adam Pearce and Irippuge Milinda Perera and Venky Ramachandran and Karthik Raman and Thomas Roessler and Izhak Shafran and Tomer Shekel and Charlotte Stanton and Jacob Stimes and Mimi Sun and Gregory Wellenius and Masrour Zoghi},
journal= {arXiv preprint arXiv:2009.01265},
year = {2020}
}