Google Inc.

Evgeny Skvortsov

Sheng Ma

James Robert Koehler

Craig Wright

Benjamin Kreuter

Arthur Asuncion

Laura Grace Book

Jiayu Peng

Xichen Huang

Carl Fredrik Hubinette

Gang Wang

Mariana Raykova

Joseph Sean Cahill Goodknight Knightbrook

Zhe Wang

Lukas Rutishauser

Thomas A Vaughan

Shayan Mirjafari

Shen-fu Tsai

Preston Wooju Lee

Wei-Ting Liao

Jeffrey Wilhelm

William George Kahn Bradbury

Andreas Ulbrich

Zhaosheng Bao

Stuart Kendrick Harrell

Yip Man Tsang

Lu Zhang

Taylan Yildiz

Cuizhu Shi

Zachary Charles Frazier

Mariana Petrova Raykova

Jeffrey Wilhem

Benjamin R Kreuter

Systems and methods for generating min-increment counting bloom filters to determine count and frequency of device identifiers and attributes in a networking environment are disclosed. The system can maintain a set of data records including device identifiers and attributes associated with device in a network. The system can generate a vector comprising coordinates corresponding to counter registers. The system can identify hash functions to update a counting bloom filter. The system can hash the data records to extract index values pointing to a set of counter registers. The system can increment the positions in the min-increment counting bloom filter corresponding to the minimum values of the counter registers. The system can obtain an aggregated public key comprising a public key. The system can encrypt the counter registers using the aggregated shared key to generate an encrypted vector. The system can transmit the encrypted vector to a networked worker computing device.

Secure multi-party reach and frequency estimation

Systems and methods for detecting a collision when combining a first encrypted data structure and a second encrypted data structure are disclosed. The system can receive the first encrypted data structure representative of a first plurality of registers. Each register in the first plurality of registers can have an encrypted fingerprint value, and an encrypted register identifier value. The system can receive the second encrypted data structure representative of a second plurality of registers. The system can calculate a first sum associated with a first register of the first plurality of registers based on the fingerprint value of the first register. The system can calculate a second sum associated with a second register of the second plurality of registers. The system can determine a validity bit associated with the collision based on a comparison of the first sum and the second sum.

Cryptographic computation techniques for multi-party reach and frequency

This disclosure relates to systems and techniques that can be implemented by content platforms to optimize (a) demographic-based digital component distribution used to categorize each user into a particular demographic so as to appropriately target that user for purposes of maximizing the efficacy of digital components shown to that user, and (b) demographic reporting used to report to digital component providers the effectiveness of the digital component.

Privacy preserving machine learning for content distribution and analysis

The present disclosure is directed to automatically generating estimation reports of an online audience. In particular, the system can include a server having a first dataset with a plurality of dimensions. Additionally, the system can receive, from a client device, a query request to estimate the online audience of the content item. The system can process, using a first machine-learned model, the query request to generate a configuration file associated with the first dataset. Moreover, the system can present, on a user interface of a client device, the configuration file. Furthermore, the system can receive a user input in response to the presentation of the configuration file and retrieve data from the first dataset based on the configuration file and the user input. Subsequently, the system can generate, using the first machine-learned model, the report based on the retrieved data from the first dataset.

Intuitive data retrieval and visualization with large language models and logic programming

Systems and methods for the meta-estimation of data structures representing identifiers are disclosed. The system maintain one or more data records comprising one or more identifiers and one or more attributes. Using the data records, the system can generate a first data structure, such as a probabilistic data structure, that represents the plurality of data records. The first data structure can have a plurality of registers. The system can identify a subset of the plurality of registers that are equal to a predetermined value, and generate a second data structure that represents the subset of the plurality of registers. The system can then store the second data structure as a meta-estimation of the first, and can utilize the second data structure in further processing operations.

Meta-estimation of bloom filters with vectors of counts

Systems and methods for fast merging of panelist activity are disclosed. The system can maintain a plurality of panelist identifiers each stored with a respective plurality of offline content events, and identify, from the plurality of panelist identifiers, a subset of panelist identifiers that are stored with a respective offline content event that matches a target offline content event. The system can map each of the subset to a respective plurality of unique identifiers corresponding to virtual devices having virtual device attributes. The system can reduce, for each of the subset, the respective plurality of unique identifiers to a sketch that represents the respective plurality of unique identifiers. The system can combine the sketch of each of the subset of panelist identifiers into an aggregated sketch, and transmit the aggregated sketch to a computing device for analysis with an aggregated sketch representing online content events.

Systems and methods for cross media reporting by fast merging of data sources

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, that facilitate resource and space efficient analysis of large scale datasets. Methods include obtaining activity data for objects in a dataset. For each data item in the dataset, a hashed parameter having a binary representation is generated using an identifier for the object. A register is identified from among a set of registers based on the hashed parameter. A determination is made that the hashed parameter for the object contributes to an aggregation amount that specifies a number of occurrences of the object in the dataset. Based on this determination, an aggregation amount stored in the register is updated. Based on aggregation amounts stored in the set of registers, a reporting output is generated that provides an aggregate distribution of the objects in the dataset based on the activity data for the objects.

Optimizing large scale data analysis

A method for efficiently translating logic programming language code into structured query language (SQL) code that includes receiving input written in a logic programming language (LPL) that indicates a function that is defined in the LPL, but is not defined in a target SQL used to access a dataset, translating the input written in the LPL into the target SQL to generate translated SQL code, including transforming the function not defined in the target SQL into code for a series of functions that (i) are defined in the target SQL and (ii) together achieve a result of the function not defined in the target SQL, and executing the translated SQL code to achieve a result of the function that is not defined in the SQL, including executing the code for the series of functions that together achieve the results of the function that is not defined in the target SQL.

Transforming logic programming language code into structured query language code

Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium, including a method for discovering unique entities over multiple devices. A virtual pool of entities is created and divided into subpools, each including fewer than all entities. Subpools are subdivided into delta pools. Cookies are recorded for each delta pool when the particular portion of content is presented to or accessed by entities in the delta pool. Recorded cookies are divided into cookie types based on cookie characteristics. Machine learning and statistical analysis algorithms are used to automatically determine sizes of delta pools and probabilities of each cookie type being classified as belonging to particular delta pools. Virtual entities are assigned from the virtual pool to each of the recorded cookies that were recorded when the particular portion of content was presented. A number of unique entities that accessed the particular portion of content is determined.

Discovery of unique entities across multiple devices

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining a number of unique viewers of a content item. In one aspect, a method includes receiving an identifier specifying characteristics of a user associated with a device that was used to present a given content item. A user category is identified for the received identifier based on the characteristics. Each user category of a set of user categories include a respective number of pseudo-users. The received identifier is assigned to a given pseudo-user in the identified user category based on a non-linear distribution of identifiers per actual user in the identified user category. A number of unique users, including the given pseudo-user, that have viewed the given content item is determined based on a total number of pseudo-users that have been assigned an identifier corresponding to presentation of the given content item.

Determining a number of unique viewers of a content item

A system and computer-implemented method for determining an estimated number of unique visitors to a network location from a geographical area. A non-linear approximation is utilized to determine the estimated number of unique visitors to the network location. The non-linear approximation is based on at least the estimated number of users within the geographical area, the estimated number of unique user identifiers within the geographical area, and the number of unique user identifiers from the geographical area that are observed at the network location.

Determining unique vistors to a network location

Growing community of inventors

Kirkland, WA, United States of America

Evgeny Skvortsov