Intuit, Inc.

Oracle International Corporation

Blue Kai, Inc.

Guohan Gao

Frank Wisniewski

Tobias Ullrich

Lars Pfannenschmidt

Shabbir Y Rajgara

Thorsten Kruger

Kendra Mariko Chen

Man Chan

Abe Taha

Aleksandar Mastilovic

Thorsten Krüger

Tharathorn Joy Rimchala

Shir Meir Lador

Sameeksha Khillan

Peter Lee Frick

Lucian Vlad Lita

Michael David Bigby

Kurt Peterson

Todd Enerson

Maojin Jiang

Aspects of the present disclosure provide techniques for image-based document search. Embodiments include receiving an image of a document and providing the image of the document as input to a machine learning model, where the machine learning model generates separate embeddings of a plurality of patches of the image of the document and the machine learning model generates an embedding of the image of the document based on the separate embeddings of the plurality of patches. Embodiments include determining a compact embedding of the image of the document based on applying a dimensionality reduction technique to the embedding of the image of the document generated by the machine learning model. Embodiments include performing a search for relevant documents based on the compact embedding of the image of the document. Embodiments include performing one or more actions based on one or more relevant documents identified through the search.

Image-based document search using machine learning

A method, system, and computer program product for managing user identification codes in an interne advertising environment. One aspect implements a system including a database engine to store a plurality of signals comprising characteristics and/or values received from a user device. A user ID generator calculates collision statistics and/or fragmentation statistics to form a first mapping function that is in turn used to generate a plurality of identification codes based at least in part a first set of selected signals. A calibration module produces measurements determined from collision quantities and/or fragmentation quantities using the first mapping function, wherein the measurements are determined by comparing the plurality of identification codes to entries in a known ID database. A sequencing module generates updated sequences of mapping functions.

Tunable statistical ids

A method, system, and computer program product for managing user identification codes in an internet advertising environment. One aspect implements a system including a database engine to store a plurality of signals comprising characteristics and/or values received from a user device. A user ID generator calculates collision statistics and/or fragmentation statistics to form a first mapping function that is in turn used to generate a plurality of identification codes based at least in part a first set of selected signals. A calibration module produces measurements determined from collision quantities and/or fragmentation quantities using the first mapping function, wherein the measurements are determined by comparing the plurality of identification codes to entries in a known ID database. A sequencing module generates updated sequences of mapping functions.

Tunable statistical IDs

Certain aspects of the present disclosure provide techniques for updating database records while maintaining accessible temporal history. One example method generally includes receiving a request, at a database, to select an instance of a record from the database at a specific point in time and reading the instance of the record from a snapshot of the database. The method further includes loading one or more deltas associated with the record from the database and chronologically applying the one or more deltas to the instance of the record to create the instance of the record. The method further includes returning the instance of the record, determining that the request has made a percentage of recent requests exceed a threshold for requests for most-current data and creating a new snapshot of the database.

Updating database records while maintaining accessible temporal history

The disclosed embodiments provide a system that manages synchronization between a profile store and a source of truth that is used to update the profile store via a real-time link. During operation, the system obtains a first set of attributes from a profile for an entity in the profile store. Next, the system obtains a second set of attributes for the entity from the source of truth. The system then compares the first and second sets of attributes to detect and correct mismatches between the first and second sets of attributes, wherein the corrected mismatches improve subsequent use of the profile in the profile store by the set of remote offerings. Finally, the system varies a rate of comparing the first and second sets of attributes for the mismatches based on one or more execution conditions associated with the source of truth.

Feedback loops for managing profile store synchronization issues

The disclosed embodiments provide a system that manages synchronization between a profile store and a source of truth that is used to update the profile store. During operation, the system obtains a request to compare a first set of attributes from a profile for an entity in the profile store with a source of truth. During a period for delaying processing of the request, the system reduces a load on the source of truth by ignoring additional requests to compare the profile with the source of truth. After the period has passed, the system obtains a first set of attributes from the profile and a second set of attributes for the entity from the source of truth. The system then compares the first and second sets of attributes to detect and correct mismatches between the first and second sets of attributes.

Intelligent batching of verification requests for profile stores

The disclosed embodiments provide a system that manages synchronization between a profile store and a source of truth that is used to update the profile store via a real-time link. During operation, the system obtains a first set of attributes from a profile for an entity in the profile store and a second set of attributes for the entity from the source of truth. The system then compares the first and second sets of attributes for mismatches between the first and second sets of attributes. When a mismatch between the first and second sets of attributes is found, the system corrects the mismatch by modifying the first set of attributes in the profile store with one or more values from the second set of attributes to improve subsequent use of the profile in the profile store by the set of remote offerings.

Managing synchronization issues between profile stores and sources of truth

The disclosed embodiments provide a system that manages synchronization between a profile store and a source of truth that is used to update the profile store via a real-time link. During operation, the system selects a profile for an entity in the profile store for use in verifying the synchronization between the profile store and the source of truth based on an access pattern associated with the profile store. Next, the system obtains a first set of attributes from a profile for an entity in the profile store and a second set of attributes for the entity from the source of truth. The system then compares the first and second sets of attributes to detect and correct mismatches between the first and second sets of attributes, wherein the corrected mismatches improve subsequent use of the profile in the profile store by the set of remote offerings.

Intelligent verification of updates to profile stores from sources of truth

The disclosed embodiments provide a system that manages synchronization between a profile store and a source of truth that is used to update the profile store via a real-time link. During operation, the system obtains a request to compare a first set of attributes from a profile for an entity in the profile store with a source of truth. During a pre-specified period for delaying processing of the request, the system reduces a load on the source of truth by ignoring additional requests to compare the profile with the source of truth. After the pre-specified period has passed, the system obtains a first set of attributes from the profile and a second set of attributes for the entity from the source of truth. The system then compares the first and second sets of attributes to detect and correct mismatches between the first and second sets of attributes.

The disclosed embodiments provide a system for updating database records while maintaining accessible temporal history. The system operates by receiving a request, at a database, to select a specific instance of a record from the database at a specific point in time. In response to the request, the system reads an instance of the record from a snapshot of the database, wherein the snapshot of the database was made prior to the specific point in time. Next, the system loads one or more deltas associated with the record from the database, wherein each delta in the one or more deltas comprises the difference between a new state of the record and a prior state of the record. The system then chronologically applies the one or more deltas to the instance of the record to create the specific instance of the record. Finally, the system returns the specific instance of the record.

A system, method, apparatus, and processor readable storage media are described for real-time computation of statistics between audiences including, but not limited to, correlations, comparative metrics, and/or other statistical analysis for the audiences. Embodiments enable a user to specify a query that defines a specific audience for analysis in real-time. A past time period may be determined that indicates the amount of historical data that is to be analyzed to generate the statistical comparisons and/or correlations. Employing the user-specified query, statistical comparisons and/or correlation may be provided, based on an analysis of retrieved historical audience data for the time period. Embodiments may also enable a user to generate new statistics in real-time through modification of the audience in one or more iterations. Embodiments may also enable a user to filter, sort, and search through audiences by setting desired statistical boundaries for the resulting audience set.

Growing community of inventors

San Jose, CA, United States of America

Guohan Gao