Google Inc.

Aiyou Chen

Tushar Udeshi

James Robert Koehler

Jeffrey David Oldham

Swaminathan Krishnamurthy

Yueqing Wang

Nicolas Remy

Yunting Sun

David Chan

Yuxue Jin

Ivan Ordonez

David Paul

Michael Perry

Timothy C Au

Kevin Benac

Timothy Chun-Wai Au

Techniques for preparing datasets for geo experiments and improving accuracy of geo experiments are presented herein. The system can access a dataset of a plurality of geographic pairs. Additionally, the system can calculate a first outcome estimate based on a difference in response data and a difference in input data for a first geographic pair. Moreover, the system can calculate a plurality of experimental uncertainty estimates associated with the plurality of geographic pairs during an experimental time interval. The system can access historical data associated with the plurality of geographic pairs. Furthermore, the system can determine a beta value and a trim rate that reduces a sum of the plurality estimates. Subsequently, the system can remove, based on the first outcome estimate and the beta value, the first geographic pair from the plurality of geographic pairs to generate the first subset of geographic pairs.

Geographic dataset preparation and analytics systems

Systems, methods and computer-readable storage media utilized to prepare datasets for geo experiments. One method includes receiving one or more input parameters. The method further includes extracting, from the data, training data. The method further includes calculating a difference in input data and a difference in response data of the training data. The method further includes determining a first plurality of geographic pairs. The method further includes extracting, from the data, evaluation data. The method further includes separating each geographic pair of the first plurality of geographic pairs into a treatment region or a control region for a plurality of simulations of a plurality of different simulation subsets for each of a plurality of different subsets of geographic pairs. The method further includes calculating a plurality of uncertainty estimates. The method further includes selecting a first subset of geographic pairs and providing the selected subset of geographic pairs.

Geographic dataset preparation system

Systems and methods for assigning word fragments to lines of text in optical character recognition (OCR) extracted data can include at least one processor obtaining a plurality of word fragments from OCR generated data associated with an image. The at least one processor can determine vertical coordinates of each of the word fragments in the image. The at least one processor can cluster the plurality of word fragments into one or more clusters of word fragments based on the vertical coordinates of the plurality of word fragments. The at least one processor can assign each word fragment of a respective cluster to a corresponding text line based on the clustering.

System and methods for assigning word fragments to text lines in optical character recognition-extracted data

Systems, methods, and computer-readable storage media that may be used to generate causal models and calculate a selection bias in mixed media. In some embodiments, the selection bias calculation is in search sponsored content in the context of mixed media modeling. In some embodiments, a method for search bias correction is based on the back-door criterion from causal inference.

Selection bias correction for paid search in media mix modeling

Systems and methods of directed item consumption recommendations are disclosed which include generating, with a server, empirical transition matrix data that includes row data for a first item and column data for a second item, and an entry in the empirical transition matrix data for a number of users that acquire the second item after the first item, generating, with the server, metadata transition matrix data by partitioning items for each item metadata type, setting a uniform transition probability for all items in the partition, and summing the uniform transition probabilities across all metadata types, generating, with the server, transition probability matrix data by multiplying the metadata transition matrix data with a weight parameter, adding the result to the empirical transition matrix data, and normalizing each row, and providing item recommendations to a user computing device communicatively coupled to the server according to the generated transition probability matrix data.

Systems and methods of providing recommendations by generating transition probability data with directed consumption

Growing community of inventors

Palo Alto, CA, United States of America

Aiyou Chen