Microsoft Technology Licensing, LLC

Yi Li

Qifa Ke

Ming Gang Liu

Kang Li

Jian Qiao Sun

Krishnaram Kenthapadi

Ganesh Venkataraman

Victor Chen

Yiping Zhou

Alexander Coll

Richard Jungiang Qian

Zhengdong Lv

Yohn Cao

Lei Zhang

Wei-Ying Ma

Heung-Yeung Shum

Fang Wen

Rolando Jimenez Salgado

Rui Hu

Yan Ke

Daniel Marantz

Ashish Consul

Krista Bendig

Xiao Wu

Xin-Jing Wang

Yanfeng Sun

Daniel J McAllister

Nathan Robert Penner

Yu-ting Kuo

Humberto Lezama Guadarrama

Mark Peterson

Travis Eby

Gabriel J Hall

Gary Randall Ratterree

Zhong Wu

Javier Garcia Flynn

Jay R Rathi

Dorrene Elizabeth Brown

Lei Tan

Yanbiao Zhao

Raul Clouse

Mauricio Ordonez

Pin Lu

Jeffrey Todd Cohen

Shreedhar Thirugnanasambantham

Dmitry Pugachev

Sean Colin Wiser

Mahesh Kumar Nayak

Robert Charles Ward

Gregory Mark Hewett

Brian Drew Lehr

Jianyong Xiao

Xun Kang

Yanhong Zhai

Richard Oian

Hong Gao

Sameera Mahesh Padhye

Vicki Liu

Felicia Werchan

Longfei Zhang

Embodiments of the disclosed technologies include receiving, from a first user device, a first user interface event that includes an upload of a media asset but does not indicate an intended usage for the media asset, determining an intended usage based on a second user interface event, mapping the intended usage to a usage-based specification, converting the media asset to a usage-specific version of the media asset by applying the usage-based specification to the media asset, and initiating a distribution of the usage-specific version of the media asset instead of the media asset to at least one second user device different from the first user device.

Contextual creation of distributable assets

Internet bound traffic that is being routed through a particular interface within an edge router can, under some circumstances, be moved to a different interface. In some situations, one or more Internet protocol (IP) flows can be moved to a different interface within the same edge router. In other situations, one or more IP flows can be moved to a different interface within a different edge router. Decisions to move Internet traffic to a different interface can be based on information about interfaces within the various edge routers in a cloud computing provider's network, information about IP flows that are being sent through the various interfaces, peering capacity information and information about the latency associated with particular Internet paths. Some or all of the information that is used in connection with traffic steering decisions can be based on real-time measurements.

Dynamic egress traffic steering for large scale cloud network

Representative embodiments allow private user created content to be submitted for a query so that a user can search from outside a corporate firewall and retrieve both private user created content and web content. The private user created content and web content are presented in the same user interface. Embodiments allow users to create content and keep a private content index that allows such content to be retrieved with an appropriate query along with public web content. Embodiments allow feedback on the user content to be submitted. Feedback can be used for quality control and other purposes. Some user created content has an expiration date after which the content will not be retrieved. Some embodiments place the private index within a compliance boundary and results are returned from within the compliance boundary directly to a browser so the search system has no access to the private data.

Private content in search engine results

Data presentation functionality associated with a collaboration database may be provided. A presentation of a table in conjunction with a collaboration database may be detected. At least one field in the table that includes people type data may also be detected. Initial letters associated with a person or an entity may be detected as input into the at least one field. A type of entry of the initial letters may be detected. A local data source may be searched to identify matching entries. Suggestions may then be displayed. If one of the suggestions is selected, additional information associated with the selected suggestion may be displayed on a contact card in a vicinity of the at least one field. An interactive functionality associated with the additional information through the contact card may be provided upon selection of the at least one field when the table is presented.

Providing data presentation functionality associated with collaboration database

The disclosed subject matter involves identifying clusters and segments of a population of data for use in a recommendation service. Clusters of members or items are formed, where the clusters, or partitions are close to being equal in size. Items are distributed based on similarities identified with matrix factorization. The items are formed into clusters based on the similarities and the clusters are used in training of a generalized linear mixed model treating the clusters as random-level effects. The trained model may be used in the recommendation service. Other embodiments are described and claimed.

Member clustering with equi-sized partitions

The disclosed subject matter involves identifying clusters and segments of a population of data for use in a recommendation service. Clusters of members or items are formed, where the clusters, or partitions are close to being equal in size, items are distributed based on similarities identified with matrix factorization. A matrix used in the matrix factorization is customized based on the recommendation type. The items are formed into clusters based on the similarities and the clusters are used in training of a generalized linear mixed model treating the clusters as random-level effects. The trained model may be used in the recommendation service. Other embodiments are described and claimed.

Efficient recommendation services

Web search results and enterprise search results are combined into a single presentation. A search query is received from a user associated with a tenant of an enterprise cloud computing platform. Web search results are retrieved based on the search query. The Web search results are ranked and sent to the search engine user interface. An intent of the search query is determined. It is determined that the search query is related to the tenant's enterprise data. An instruction containing the search query keywords and intent is sent to the search engine user interface, and is configured to cause the search engine user interface to submit the keywords intent to an enterprise search engine within the enterprise cloud computing platform. A content template is communicated to the search engine user interface. The content template is configured to display enterprise search results received at the search engine user interface.

Blending enterprise content and web results

An interface mechanism is provided for developers to bundle together a set of extensions (or add-ins) to a productivity suite allowing end users and system administrators to manage their lifecycle in a unified way. Aspects such as discovery, acquisition, authorization deployment, management, updates, and similar ones may be addressed for the bundle as opposed to each of the different extensions (for example extensions for different versions of the same application such as desktop version, web application version, mobile version, etc.). Bundles may also provide a means to match specific functional sets or commercial offerings to specific user or customer profiles.

Extensibility bundles for a cloud and devices suite

An entity-based search system is described herein that detects and recognizes entities in Internet-based content and uses this recognition to organize search results. The system associates one or more entity identifiers with a web page and stores this information as metadata of the page in a search engine index. This metadata will enable entity-based queries as well as rich data presentations in a search engine result page (SERP), including grouping results by entities, filtering results by one or more particular entities, or re-ranking search results based on user preference of entities. Thus, the entity-based search system allows users to identify a particular entity the user is interested in finding, and to receive search results directly related to that entity.

Knowledge-based entity detection and disambiguation

Image descriptor identifiers are used for content-based search. A plurality of descriptors is determined for an image. The descriptors represent the content of the image at respective interest points identified in the image. The descriptors are mapped to respective descriptor identifiers. The image can thus be represented as a set of descriptor identifiers. A search is performed on an index using the descriptor identifiers as search elements. A method for efficiently searching the inverted index is also provided. Candidate images that include at least a predetermined number of descriptor identifiers that match those of the image are identified. The candidate images are ranked and at least a portion thereof are presented as content-based search results.

Content-based image search

Techniques for identifying near-duplicates of a media object and associating metadata of the near-duplicates with the media object are described herein. One or more devices implementing the techniques are configured to identify the near duplicates based at least on similarity attributes included in the media object. Metadata is then extracted from the near-duplicates and is associated with the media object as descriptors of the media object to enable discovery of the media object based on the descriptors.

Associating media with metadata of near-duplicates

Methods and computer-storage media having computer-executable instructions embodied thereon that facilitate refining query results using visual cues are provided. Query results are determined in response to an indication of a user query. One or more groups of query results are generated from the query results based on categories of query results that share similar features. Visual cues are associated with each of the query result groups. Visual cues, in association with query result groups, are presented to a user. Query results associated with a selected visual cue may be presented to a user. A refined user query may be generated based on a selected visual cue.

Visual-cue refinement of user query results

Systems, computer-readable media, and methods for generating search engine results pages are provided. A user provides a search engine with one or more query terms. The query terms may be associated with an intent, e.g. product entity, store entity, person entity. The search engine classifies the query and identifies search results that correspond to the query. The search results are grouped based on intents associated with the query. A graphical user interface displays the grouped on search results and allows the user to modify the groupings. The graphical user interface is also updated with entity or entity attributes corresponding to the intents used to group the search results.

Disambiguating intents within search engine result pages

Multiple search indexes can be served from a common set of resources. Instead of requiring a processor to be dedicated to serving a single search index, a processor can provide responsive documents for search queries that are based on different ranking algorithms and/or different sets of documents.

Serving multiple search indexes

Growing community of inventors

Issaquah, WA, United States of America

Yi Li