Amazon Technologies, Inc.

A9.com, Inc.

Rahul H Bhagat

Michael Lee Brundage

Daniel R Parshall

Jody David Biggs

Brian Cameros

Srikanth Thirumalai

Scott Allen Mongrain

Logan Luyet Dillard

Grant Michael Emery

Adam Callahan Sanders

Kemal Oral Cansizlar

Ravendar Lal

Benjamin Thomas Cohen

Taruja Sanjay Parande

Yongwen Liang

Vaclav Petricek

Srevatsan Muralidharan

Eli Glen Rosofsky

Michael David Quinn

Vishnu Narayanan

Systems and methods are disclosed for optimizing responses to queries. Analyses of user interactions and other behaviors can lead to association of queries with signals, including ASINs and other product descriptions. The associations can be algorithmically graphed and analyzed on a disaggregated basis and individually weighted to improve search recall and reduce the risk of returning defective search results. Machine learning techniques can further optimize the associations and/or the search results.

Weighted behavioral signal association graphing for search engines

Disclosed are various embodiments for predicting repeated behavior by an entity with respect to a particular item. In some embodiments, an entity requesting results to a search query may be redirected to an alternative network page associated with the particular item instead of a search results page according the expectation of repeated behavior.

System and method for personalized network content generation and redirection according to repeat behavior

User interactions with presented content can be tracked and a corresponding user behavior identified. An enhancement trigger can be detected by comparing the user behavior with a pattern of behavior associated with the enhancement trigger. A modified experience can be provided at the user device. The modified experience can include a user interface for interacting with a chat bot.

Search result refinement and item information exploration

Disclosed are various embodiments for personalized network content generation and redirection according to time intervals between repeated instances of behavior. A computing device determines an expectation of a specific entity repeating a behavior associated with an item based at least in part on a time interval between prior occurrences of the behavior by the specific entity. The computing device then selects the item from a plurality of items based at least in part on the expectation of the specific entity repeating the behavior. Subsequently, the computing device generates a network page including a recommendation to repeat the behavior. Next, the computing device sends the network page to a client device associated with the specific entity.

Personalized network content generation and redirection according to time intervals between repeated instances of behavior based on entity size

Product information may be utilized to create a translation dictionary. The translation dictionary may then be utilized to translate search queries from a foreign language to the primary language that is utilized to provide an online e-commerce marketplace. The translated search queries may then be utilized to perform a search of a product catalog maintained by the online e-commerce marketplace. The translation dictionary created with the product information might also be utilized to translate resources, such as Web site resources, to verify the translation of the resources, and to perform other functionality.

Foreign language translation using product information

Disclosed are various embodiments for predicting repeated behavior by an entity with respect to a particular item. In some embodiments, a network page including a recommendation for the particular item may be generated according to an expectation of repeated behavior. In some embodiments, an entity requesting results to a search query may be redirected to an alternative network page associated with the particular item instead of a search results page according the expectation of repeated behavior.

Personalized network content generation and redirection according to repeat behavior

Various embodiments are included for adjusting a search result user interface when a foreign language search query is detected. A search query associated with a search request is obtained. A language of the search query is identified. Search results are obtained from a site corresponding to the language. Additionally, a translation of the search query is obtained if allowed for a corresponding product category.

Adjusting search result user interfaces based upon query language

Various embodiments for adjusting a search result user interface when a foreign language search query is detected. A search query associated with a search request is obtained. A language of the search query is identified. An search result user interface is generated using a foreign language template when the language varies from an expected language. The search result user interface includes results based on the foreign language search query.

The present technology may identify item category affinities by identifying a plurality of classifications of an item. An accuracy of the plurality of classifications relative to one another for the item may be identified. A category affinity of the item may be determined based on the accuracy of the plurality of classifications relative to one another.

Video classification

A user supplied search query is received. Historical search queries are identified that are relevant and/or similar to the user supplied search query. The identified historical queries are then ranked according to various factors. Suggested search queries that are relevant to the user supplied query are identified.

Generating suggested search queries

Disclosed are various embodiments of adjusting a search result user interface when a foreign language search query is detected. A search query associated with a search request is obtained. A language of the search query is identified. An alternative search result user interface is generated when the language varies from an expected language. Additionally, an alternative search query can be generated that is based at least in part upon the originally submitted search query.

The use of graph-based semantic analysis with respect to items and tags may enable the discovery of the characteristics of items. A tag collection component may initially obtain a corresponding set of tags for each item of multiple items. A graph generation component may then generate a graph that includes a corresponding item node for each item and a corresponding tag node for each tag cluster of tags. Each item node in the graph may be connected to each of one or more tag nodes by a respective edge. Subsequently, following assignment of a label to each tag node, a graph evaluation component may execute a random walk algorithm on the graph. The execution of the random walk algorithm may provide a corresponding ranked list of tags for each item or a corresponding set of correlated tags for each tag.

Growing community of inventors

Bellevue, WA, United States of America

Rahul H Bhagat