Ebay, Incorporated

Paypal, Inc.

Nicholas Whyte

Davide Libenzi

Robinson Piramuthu

Richard Dean Henderson

Mohammadhadi Kiapour

Shuai Zheng

Padmapriya Gudipati

Swati Agarwal

Timothy Keefer

Ashmeet Singh Rekhi

Amit Basu

Md Atiq Ul Islam

Giridharan Iyengar

Thomas Pan

Pradeep Sankaranthi

Anand Lakshminath

Guanglei Song

Prasad Sriram

Bryan Allen Plummer

Andrew John Melnick

Alberto Ordonez Pereira

Ravindra Surya Lanka

Raffi Tutundjian

Mel Sun

Jagadish Nallapaneni

Carson Hoffacker

Gang He

Scot Fraser Hamilton

Snezana Sahter

Suchitra Ramesh

Shane Lin

John Weigel

Fujun Zhong

Niaz Ahamed Khaja Nazimudeen

Kenneth Clark Crookston

Rekha Patel

Satya Seethasridhar

Parikshit Arora

Techniques and systems are described that leverage computer vision as part of search to expand functionality of a computing device available to a user and increase operational computational efficiency as well as efficiency in user interaction. In a first example, user interaction with items of digital content is monitored. Computer vision techniques are used to identify digital images in the digital content, objects within the digital images, and characteristics of those objects. This information is used to assign a user to a user segment of a user population which is then used to control output of subsequent digital content to the user, e.g., recommendations, digital marketing content, and so forth.

Computer vision, user segment, and missing item determination

In various example embodiments, a system and method for versioning data in an unstructured data store are presented. The method includes generating a first record in the unstructured data store based on a version of a second record in the unstructured data store. The method further includes initializing a parent version identifier of the first record based on the version of the second record. The method further includes comparing the parent version identifier of the first record to a version identifier of the second record. The method further includes determining whether the first record is consistent based on the comparison.

Providing custom read consistency of a data object in a distributed storage system

Computer vision and image characteristic search is described. The described system leverages visual search techniques by determining visual characteristics of objects depicted in images and comparing the determined characteristics to visual characteristics of other images, e.g., to identify similar visual characteristics in the other images. In some aspects, the described system performs searches that leverage a digital image as part of a search query to locate digital content of interest. In some aspects, the described system surfaces multiple user interface instrumentalities that include images of patterns, textures, or materials and that are selectable to initiate a visual search of digital content having a similar pattern, texture, or material. The described aspects also include pattern-based authentication in which the system determines authenticity of an item in an image based on a similarity of its visual characteristics to visual characteristics of known authentic items.

Computer vision and image characteristic search

Methods and systems to build and utilize a search infrastructure are described. The system generates, by a first indexing subsystem, full indexes of data and sends the full indexes to a plurality of query node servers via a distribution system. The system generates, mini-indexes associated with the full indexes and sends the mini-indexes to the plurality of query node servers. The system causes the plurality of query nodes to update indexing information based on the full indexes and the mini-indexes.

Next generation near real-time indexing

Computer vision for unsuccessful queries and iterative search is described. The described system leverages visual search techniques by determining visual characteristics of objects depicted in images and describing them, e.g., using feature vectors. In some aspects, these visual characteristics are determined for search queries that are identified as not being successful. Aggregated information describing visual characteristics of images of unsuccessful search queries is used to determine common visual characteristics and objects depicted in those images. This information can be used to inform other users about unmet needs of searching users. In some aspects, these visual characteristics are used in connection with iterative image searches where users select an initial query image and then the search results are iteratively refined. Unlike conventional techniques, the described system iteratively refines the returned search results using an embedding space learned from binary attribute labels describing images.

Computer vision for unsuccessful queries and iterative search

Methods and systems to build and utilize a search infrastructure are described. The system generates index information components in real-time based on a database that is time-stamped. The system updates index information at a plurality of query node servers based on the index information components. A query engine receives a search query from a client machine and identifies search results based on the query and the index information. The system communicates the search results, over the network, to the client machine.

Systems and methods to build and utilize a search infrastructure

Methods and systems to build and utilize a search infrastructure are described. The system generates index information, including document updates and indexes. The system receives event notifications as the document updates are received and accumulates the index information until published. A query engine receives a search query from a client machine and identifies search results based on the query and the index information. The system communicates the search results, over the network, to the client machine.

Systems and methods to configure a query language using an operator dictionary are described. The system receives, from a first user, a search query to search for document information in a database. The system determines that terms of the search query from the first user are in accordance with dictionary information that corresponds to the first user. The dictionary information includes a first list of compatible query operators. The system applies the search query to the index data structure in the database to identify the document information. The system sends an electronic interface to the first user that includes the document information.

System and methods to configure a query language using an operator dictionary

A method and a system to configure a query language using an operator dictionary are provided. Dictionary information that corresponds to a first user is generated. The dictionary information includes a first list of compatible query operators and a description of compatible data types that correspond to each operator in the first list of query operators. A search query to search for document information in a database is received from a first user. The database includes an inverted index. Terms of the search query from the first user are determined to be in accordance with the dictionary information that corresponds to the first user. The search query is applied to the inverted index in the database to identify the document information. The document information is sent to the first user.

An input query can be spell checked by first parsing the input query into one or more tokens. The one or more tokens are then compared against spell storage tables to derive alternative tokens. A spell cursor is launched, with the spell cursor calculating scores for permutations of the tokens and alternative tokens, and the scores representing a likelihood that a user intended to use the permutation as the input query. A memory table may then be formed including the permutations and scores. The memory table can then be passed to a column cursor, the column cursor designed to form an output table of results based on queries to a database using the permutations.

Spell check using column cursor

Growing community of inventors

San Jose, CA, United States of America

Nicholas Whyte