Microsoft Technology Licensing, LLC

Derrick Leslie Connell

Brian W MacDonald

Sanaz Ahari

Sandy Wong

Paul Ronald Ray

Mira Lane

Kang Li

Yinghua Yao

Xavier Legros

Jun Yin

Javier Garcia Flynn

Kevin L Haas

Andrew Shuman

Ravi Soin

Evan Malahy

Heung Yeung Shum

Marek Latuskiewicz

Kim Vlcek

William Henry Gates, Iii

Hugh Evan Williams

Gopi Prashanth Gopal

Qi Lu

Darren Arthur Shakib

Gaurav Sareen

Nitin Agrawal

Daniel Marantz

Milad Shokouhi

Samir Shiraz Pradhan

Andrew Peter Oakley

Jay Goyal

Neil Sharman

Christopher A Meyers

Nicholas Craswell

Scott J Beaudreau

Jeffrey R Kelleran

James P Dierkes, Jr

Brian Scott Otto

Brian Macdonald

Search results can include, not only responsive content, typically in the form of documents, but can also include information regarding responsive entities, including tasks relevant to those entities, and can include identifications of individuals from whom additional information can be obtained, or who are otherwise relevant to the search. An association between entities and documents referencing those entities is identified and maintained. Documents responsive to a user's search query are identified utilizing existing algorithmic mechanisms and such identified responsive documents are referenced to determine whether they are associated with one or more entities. Additionally, a people/query mapping is maintained associating individual people with specific queries. User actions directed to one or more people in response to a query are utilized as a feedback loop. User actions directed to one dimension of search results can be utilized to generate updated search queries providing updated search results in other dimensions.

Multidimensional search architecture

A unified experience environment supports mechanisms that collect and utilize contextual metadata to associate information in accordance with its relevance to a user's current context. An ambient data collector obtains contextual and activity information coincident with a user's creation, editing or consumption of data and associates it with such data as contextual metadata. A context generator that utilizes contextual metadata to identify relationships between data and enable the proactive presentation of data relevant to a user's current context. Proactive presentation includes a context panel that is alternatively displayable and hideable in an application-independent manner and a unified activity feed that comprises correlated data groupings identified by correlation engines, including a universal, cross-application correlation engine and individual, application-specific correlation engines that exchange information through data correlation interfaces. The context panel and unified activity feed enable users to more efficiently access data and increase their interaction performance with a computing device.

Unified presentation of contextually connected information to improve user efficiency and interaction performance

Computer-implemented methods and a computing system are provided for presenting a new user interface paradigm which allows users to explore facets related to a search query term while maintaining the original state of the search query term or staying within the confines of the user defined topic. The methods and system present facets or categories of relevant topics of a user generated search query term. The user may then explore these facets in a number of embodiments while maintaining the state of the search.

Table of contents for search query refinement

Computer-storage media, methods, and systems for improving the ability of a user to accomplish a task that is pending during a search session are provided. When a user invokes the pending task, a collection of elements that are associated with the pending task are dynamically organized into a sharable content page that may be shared with a user's contacts via a social network. The user's contacts may comment on or provide feedback related to the pending task. The feedback may be published to the sharable content page and presented to the user on a user interface.

Organizing search history into collections

Systems, methods, and computer-readable storage media for providing task-oriented query-completion suggestions are provided. A portion of a search query is received, for instance, via a query-input region or search box presented in association with a general web search domain. As the portion of the search query is received, it is determined that the received search query portion is suggestive of a task-oriented user intent. Also determined are query-completion suggestions that are associated with the task-oriented user intent of which the search query portion is suggestive. The determined query-completion suggestions are then provided to the user. As query-completion suggestions are surfaced with input of short character sequences that are associated with task-oriented user intents, users can quickly be provided the relevant, rich content for completing their intended tasks.

Task-oriented query-completion suggestions with shortcuts

Computer-readable media and methods for model-based searching to provide search results of relevance to a search topic are provided. A model describing a search framework for generating a plurality of search queries is provided. A number of search topics are identified and attributes are assigned to each search topic. Attribute data elements and values are provided to describe each to the assigned attributes for each search topic. The attributes inform the search framework in order to compile a plurality of search queries for the search topic. A search term is received from a user and a corresponding search topic is identified. The plurality of search queries are executed to obtain search results for the corresponding search topic. The search queries may be pre-executed and the search results cached in a computer memory for later retrieval. The search results are organized based on the model and are presented to the user.

Model-based searching

Computer-storage media, methods, and systems for improving the ability of a user to accomplish a task that is pending during a search session are provided. This pending task can be distilled by identifying a current searching intent associated with a recently submitted query and by determining which candidate tasks are most comparable to the current searching intent of a user. The candidate tasks are mined from a search history of the user by categorizing those navigation events that appear to share a semantically common concept. Upon performing a comparison, the candidate task that captures the user's true current searching intent is designated as the pending task. When a user invokes the pending task, a collection of elements that are associated with the pending task are dynamically organized into a sharable content page, thereby providing a collaborative experience that involves exposing the collection to contacts and accepting feedback therefrom.

Methods, systems, and media are provided for identifying and clustering queries that are rising in popularity. Resultant clustered queries can be compared to other stored queries using textual and temporal correlations. Fresh indices containing information and results from recently crawled content sources are searched to obtain the most recent query activity. Historical indices are also searched to obtain temporally correlated information and results that match the clustered query stream. A weighted average acceleration of a spike can be calculated to distinguish between a legitimate spike and a non-legitimate spike. Legitimate clusters are combined with other stored clusters and presented as grouped content results to a user output device.

Detecting spiking queries

Growing community of inventors

Bellevue, WA, United States of America

Derrick Leslie Connell