Google Inc.

Moti Yung

Gang Wang

Daniel Ramage

Michael Kleber

Josh Karlin

Charlie Harrison

Jessica Nicola Staddon

Daniel Wesley Dulitz

Pavani Naishadh Diwanji

The present disclosure provides systems and methods for content quasi-personalization or anonymized content retrieval via aggregated browsing history of a large plurality of devices, such as millions or billions of devices. A sparse matrix may be constructed from the aggregated browsing history, and dimensionally reduced, reducing entropy and providing anonymity for individual devices. Relevant content may be selected via quasi-personalized clusters representing similar browsing histories, without exposing individual device details to content providers.

Low entropy browsing history for ads quasi-personalization

A system and method for generating privacy-enhanced aggregate statistics within a social network system is provided. Data is collected and processed to gather information to generate the aggregate statistics. A threshold is assigned. The threshold includes a criterion used in making a determination on what aggregate statistic will be generated. In some embodiments, the threshold is a numerical value. In some embodiments, the numerical value, or quantitative data is then translated into qualitative descriptors. In some embodiments, noise is then added to randomize the assigned threshold. In other embodiments, noise is added to the collected data. In some embodiments, checks to guard against attacks from adversarial users are performed. Examples of indications of adversarial behavior include, but are not limited to, manipulation of profiles, continuous manipulation of affinity groups, and manipulation of preferences for one or more users. The threshold is applied and aggregate statistics are generated.

System and method for generating privacy-enhanced aggregate statistics

Growing community of inventors

New York, NY, United States of America

Moti Yung