Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 20 - Industrial Data Science: Transforming Analytics into Business Impact
Type: Topic-Contributed
Date/Time: Sunday, August 8, 2021 : 1:30 PM to 3:20 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #317387
Title: Data Science in Online Advertising: Differentially Private Reach and Frequency Estimation for Effectiveness Measurement
Author(s): Jiayu Peng*
Companies: Google Inc.
Keywords: Differential privacy; Brand measurement; Reach; Algorithm
Abstract:

Reach and frequency are two of the most important metrics in advertising management. Ads are distributed to different publishers with a hope to maximize the reach at effective frequency. Reliable cross-publisher reach and frequency measurement is called for, to assess the actual performance of branding and to improve the budget allocation strategy. However, cross-publisher measurement is non-trivial particularly under strict differential-privacy restrictions. This paper introduces the first locally-differentially-private solution in the literature to cross-publisher reach and frequency estimation. The solution consists of a family of algorithms based on a data structure called Vector of Counts (VoC). Complying with the standard definition of differential privacy, the solution prevents attackers from telling if any specific user is reached or not with a given level of confidence. The solution enjoys particularly high accuracy for the estimation between two publishers. For more than two publishers, the solution enjoys small variance, at a risk of having bias in the presence of cross-publisher correlation of user activity.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program