Reproduce “Reach Measurement, Optimization and Frequency Capping In Targeted Online Advertising Under k-Anonymity” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yuan Gao, Mu Qiao
The paper presents methods for measuring and optimizing advertising reach and frequency under k-anonymity, a privacy-preserving protocol for user data. It quantifies the trade-off between user privacy and advertising efficiency, demonstrating that modeling non-uniform user activity within anonymous groups enhances the accuracy of both measurement and optimization.
Key insights
- Introducing k-anonymity significantly increases measurement uncertainty and reduces advertising efficiency, but subsequent increases in the group size 'k' beyond an initial threshold lead to diminishing additional performance penalties.
- The proposed 'probabilistic discounting' mechanism allows for effective reach optimization and frequency capping even when individual user identities are obscured.
- Modeling non-uniform user activity patterns within k-anonymous groups, using property vectors, is crucial for accurate reach measurement and significantly improves advertising optimization performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.