Yun Zhou - Mountain View CA, US Gaurav Garg - Palo Alto CA, US Radhika Malpani - Palo Alto CA, US
Assignee:
Google Inc. - Mountain View CA
International Classification:
G06F 17/30 G06F 7/00
US Classification:
707728
Abstract:
Methods, systems and apparatus, including computer programs encoded on a computer storage medium, for determining similar queries for image searches. In one aspect, a method includes generating, for each of a plurality of search queries, a selection vector for the search query, each selection vector including a plurality of vector elements, each vector element corresponding to a unique image, and the value of each vector element being proportional to a number of selections of its corresponding unique image in response to the unique image being presented as a search result for the search query. The method further includes selecting a first search query and a second search query from the plurality of search queries and determining, from the selection vectors for the first and second search queries, a similarity measure that is a measurement of the similarity of the first search query to the second query.
In one implementation, a method is described for identifying candidate query refinements for a given query from past queries submitted by one or more users. The method includes generating from the candidate query refinements, term occurrence scores for terms in the candidate query refinements indicating frequency of the terms within the candidate query refinements. The method further includes calculating query refinement scores for the candidate query refinements of the given query using the term occurrence scores. The method further includes storing data associating the query refinement scores with the candidate query refinements for use in selecting from among the candidate query refinements for the given query.
Validating Translations Using Visual Similarity Between Visual Media Search Results
Yun Zhou - Mountain View CA, US Vanessa Hodge - San Francisco CA, US Gaurav Garg - Palo Alto CA, US Radhika Malpani - Palo Alto CA, US
Assignee:
Google Inc. - Mountain View CA
International Classification:
G06F 17/30
US Classification:
707728, 707760, 707737
Abstract:
Methods, systems, and apparatus, including computer program products, for validating candidate translations for a phrase. A plurality of candidate translations for a phrase are received. A similarity score for each of the candidate translations is calculated. Each similarity score is an estimate of visual similarity between a group of visual media search results responsive to a visual media search query with text corresponding to the phrase and a group of visual media search results responsive to a visual media search query with text corresponding to a candidate translation. One or more of the candidate translations are selected based on the similarity scores. The selected candidate translations are associated with the phrase as valid translations for the phrase.
Clustering Documents Based On Common Document Selections
Yun Zhou - San Jose CA, US Anna Dagna Majkowska - San Francisco CA, US
Assignee:
Google Inc. - Mountain View CA
International Classification:
G06F 7/00
US Classification:
707737, 707749
Abstract:
One or more server devices may receive first navigation information identifying a first set of documents that are selected after a first document is presented and second navigation information identifying a second set of documents that are selected after a second document is presented; compare the first set of documents to the second set of documents; generate a similarity score based on the comparing; determine based on the similarity score, that the first document is similar to the second document; and generate, based on determining that the first document is similar to the second document, a cluster that includes identification information identifying the first document and the second document.
A vapor compression distillation system is disclosed. A pulsed power scale prevention module is provided in a recirculation loop of an evaporation loop and in each of a plurality of pre-evaporation heat exchange loops to nucleate contaminant particles in the liquid. The nucleated contaminant particles are removed from the system by filters in communication with the pulsed power scale prevention modules and by solids collection and discharge ports provided by heat exchangers in the evaporation loop and each of the pre-evaporation heat exchange loops. The evaporation loop and the pre-evaporation heat exchange loops provide a temperature gradient having discreet temperature zones.
Determining Dynamic Interaction Condition(S) For Triggering Provision Of Related Content Interface Notification
- Mountain View CA, US Sha Hossain - Mountain View CA, US Radwa Metwali - Mountain View CA, US Yun Zhou - Mountain View CA, US
International Classification:
G06N 20/00
Abstract:
Implementations determine attribute(s) for an Internet resource; process the attribute(s) to generate predicted output; determine, based on the predicted output, interaction condition(s) for triggering provision of a related content interface notification for the Internet resource; and responsive to access of the Internet resource by a given client device, and responsive to determining the interaction condition(s): cause the given client device to render the related content interface notification in response to determining that the access of the Internet resource satisfies the interaction condition(s). In some implementations, the interaction condition(s) vary from Internet resource to Internet resource and/or can vary for a single Internet resource (e.g., based on a navigation path used in accessing the Internet resource and/or the client device used in accessing the Internet resource).
- Portland OR, US Yun Zhou - Sunnyvale CA, US Nicolas Falletto - Pessac, FR
Assignee:
Electro Scientific Industries, Inc. - Portland OR
International Classification:
C03B 33/09 C03B 33/10
Abstract:
An internal feature can be laser machined in strengthened glass sheets or panels by first laser machining a first scribe in a first surface proximate to the internal feature to be laser machined. The internal feature can be then laser machined by positioning a beam waist of a laser beam proximate to an opposite second surface by focusing the laser beam through the strengthened glass panel from the first surface. The internal feature is laser machined by repositioning the beam waist from the second surface to the first surface while removing material from a kerf surrounding the internal feature. When the laser beam waist is finally positioned proximate to the first surface material, the internal shape formed by the laser machining is easily and cleanly removed from the surrounding glass.
Thomas J. Duerig - Sunnyvale CA, US Charles J. Rosenberg - Cupertino CA, US Kunlong Gu - Belmont CA, US Samy Bengio - Mountain View CA, US Yun Zhou - San Jose CA, US
International Classification:
G06F 17/30
Abstract:
A method includes receiving a search query comprising one or more query terms, receiving image relevance models that each generate relevance measures of content feature values of images to a query, each image relevance model being a predictive model that has been trained based on content feature values of a set of training images, and each of the queries being a unique set of one or more query terms received by a search system as a query input, identifying an image relevance model for a different query that has been identified as similar to the received search query, and calculating a fractional adjustment multiplier for search results responsive to the received search query, the fractional adjustment multiplier being based on a relevance measure generated by the identified image relevance model for the different query and based on a degree of similarity between the different query and the received search query.