Microsoft Technology Licensing, LLC

Other

Vijay Mital

Darryl Ellis Rubin

Suraj T Poozhiyil

Olivier Colle

Robin Abraham

David G Green

Brian C Beckman

Andrew Douglas Reddish

John Alfred Payne

Nizam Anuar

Gary Shon Katzenberger

Andrew Robert Miller

Moe Khosravy

Liang Du

Ranjith Narayanan

Sandeep Suresh

Max Uritsky

Jason A Wolf

Amir Netz

Gueorgui Chkodrov

Christian Beaumont

Arnaud Christian Flutre

Henry Hun-Li Reid Pan

David George Green

Alexander Sasha Stojanovic

Jaideep Sarkar

Faisal Khaled Faisal Ilaiwi

Bao Quoc Nguyen

Ahmad Nizam Anuar

Anthony Wah Lee

Robert M Fries

Michael J McCormack

Oscar E Murillo

Gregory Scott Lindhorst

Arye Gittelman

Erik Rucker

Florian Voss

Vikram Padmakar Bapat

Colleen G Estrada

Radu Bogdan Gruian

Robert John Anderson

G Shon Katzenberger

Gregory Robert Prickril

Jeremy William John Newton-Smith

Chandrika G Shankarnarayan

Maria Belousova

Curtis DeSantis

George Edward Busby

Henricus Johannes Maria Meijer

Steven Miles Greenberg

Clinton Dee Covington

Neil Warren Black

Scott Timothy Gardner

Sumit Chauhan

Pavel R Karimov

R Donald Thompson, Iii

Zachary P Woodall

Saurab Nog

Ulrich Homann

Jason A Bould

David Jon Conger

Greg S Lindhorst

Maarten Mullender

Robert Coggins

Janine Crumb

Robert Scott Cooper

Robert Povey

Benjamin Neiman Hodes

Michael J Eatough

Wolfgang Werner Hilpert

Tim Getsch

James Jackson Coddington

Krasimir A Aleksandrov

Moahemmed Amine Benmouffok

Performing collaborative search engine searching. The method includes receiving user input at a user interface for performing a plurality searches on a first search engine. The method further includes receiving user input at the user interface applying one or more augmentation AI models to searches in the plurality of searches. The method further includes creating a shareable, executable package executable by one or more search engines based on the plurality of searches and the applied AI models that when executed by the search engines causes the search engines to apply the AI models to searches performed at the search engines.

Search engine functionality using shared AI models

Expanding search engine functionality using AI models. A method includes, as part of a search session, receiving user input at a search engine. One or more searches on a set of data using the user input. Search results are provided from the one or more searches to a user. Based on a history of the search session, suggestions are provided in a user interface of AI models that could be applied to expand potential search results for the search session. User input is received at the user interface selecting one or more of the suggested AI model. The one or more selected AI models are applied to expand the set of data. Search results to the user based on searching the expanded set of data.

Expanding search engine capabilities using AI model recommendations

The improved exercise of artificial intelligence by providing a systematic way for a computing system to interface with output from AI models. To do this, the computing system obtains results of an input data set being applied to an AI model. The results are then refined based upon characteristic(s) of the AI model and perhaps the input data set. Based upon characteristic(s) of the AI model and perhaps the input data set, interface element(s) are identified that can be used to interface with the refined results. The interface element(s) are then communicated to an interface element that interfaces with the refined results. The interface element(s) may include, for instance, operator(s) or term(s) that may be used to query against the refined results and/or an identification of visualization(s) that may be used to present to a user results of queries against the refined results.

Interfacing with results of artificial intelligent models

The improved exercise of artificial intelligence. Raw output data is obtained by applying an input data set to an artificial intelligence (AI). Such raw output data is sometimes difficult to interpret. The principles defined herein provide a systematic way to refine the output for a wide variety of AI models. An AI model collection characterization structure is utilized for purpose of refining AI model output so as to be more useful. The characterization structure represents, for each of multiple and perhaps numerous AI models, a refinement of output data that resulted from application of an AI model to input data. Upon obtaining output data from the AI model, the appropriate refinement may then be applied. The refined data may then be semantically indexed to provide a semantic index. The characterization structure may also provide tailored information to allow for intuitive querying against the semantic index.

Exercising artificial intelligence by refining model output

The managing of sensed signals used to sense features of physical entities over time. A computer-navigable graph of sensed features is generated. For each sensed feature, a signal segment that was used to sense that feature is computer-associated with the sensed feature. Later, the graph of sensed features may be navigated to that features. The resulting signal segment(s) may then be access allowing for rendering of the signal evidence that resulted in the sensed feature. Accordingly, the principles described herein allow for sophisticated and organized navigation to sensed features of physical entities in the physical world, and allow for rapid rendering of the signals that evidence that sensed features.

Organization of signal segments supporting sensed features

The sharing of signal segments using a semantic understanding of what is included within the signal segment. Upon selection of a physical entity (or portion thereof) that is rendered within a signal segment, the system then extracts portion(s) of signal segment(s) in which that selected physical entity appears. A representation of the extracted signal segment portion are then dispatched to recipients. Thus, portions of signal segments may be shared at the granularity of the content of what the signal segment depicts. There is wide flexibility in what representations of those signal segment portion(s) are shared.

Sharing signal segments of physical graph

Routing of communications to group member(s) where group membership is identified by physical status. A computing system detects a communication that identifies targets of the communication at least in part by physical status. The system responds by identifying at least partial membership in a group that is identified by the physical status identified in the communication, and then dispatching the communication to at least one member of the members of that group. The identity of the members of the group may change dynamically as the physical status of particular physical entities changes over time. Accordingly, a user may communicate to individuals based on physical status, rather than identify any particular individual or status group. The communication may be directed to all members of the group. Alternatively, the communication may be initially directed towards a subset of the group, with the communication being conditionally later broadened.

Communication routing based on physical status

The improved exercise of artificial intelligence by systematically refining and semantically indexing the output from AI models, so that the semantic index is highly relevant. To do this, the computing system obtains results of an input data set being applied to an AI model. The computing system then determines a refinement to apply to the obtained results. This determination may be based on one or more characteristics of the AI model and/or input data set. The determination may also be based on hints associated with that AI model, and/or learned behavior regarding how that AI model is typically used. The obtained results are then refined using the determined refinement. It is then this more relevant refined results that are semantically indexed to generate the semantic index. Thus, the semantic index represents, the more useful output from an AI model, which is semantically exposed so as to provide meaning.

Indexing refined output of artificial intelligence models

Automatic generation of a narration of what is happening in a signal segment (live or recorded). The signal segment that is to be narrated is accessed from a physical graph. In the physical graph, the signal segment evidences state of physical entities, and thus has a semantic understanding of what is depicted in the signal segment. The system then automatically determines how the physical entities are acting within the signal segment based on that semantic understanding, and builds a narration of the activities based on the determined actions. The system may determine what is interesting for narration based on a wide variety of criteria. The system could use machine learning to determine what will be interesting to narrate.

Automatic narration of signal segment

The automatic formulation of a plan for concurrent movement of physical entities within a physical space. A physical graph is used to formulate such a plan. The physical graph represents multiple physical entities that have been sensed in a physical space over time. A plan is then formulated based on an evaluation of that physical graph. Such plans are enabled by the semantic understanding of the physical space and its contents that the physical graph provides. The plan honors physical constraints of the physical space, and physical constraints of the physical entities that are moving within that physical space. The plan may be further orchestrated by communicating with the physical entities to provide instructions for movement. Then, movement is monitored to determine if the plan is being complied with. If the plan is not being complied with, further communications are made and/or an alternative plan is automatically constructed.

Automated movement orchestration

A system and method for offering suggestions to a user of a mobile computing device based on information relevant to the user and a context data. The mobile computing device has access to lists of data associated with the user. The mobile computing device also has a sensor for detecting data about the user's context. Based on the lists of information and the contextual data, the device generates a hypothesis as to information that may be of interest to the user. Using context data in conjunction with lists in this fashion focuses the system on information likely of interest, without requiring express user input. A search may be performed in accordance with the hypothesis. Based on the results of the search, one or more suggestions are then presented by the mobile computing device to the user. The user has the option to provide feedback input to the device resulting in an update of the suggestion.

System of providing suggestions based on accessible and contextual information

A data marketplace infrastructure provides a crowd sourcing solution to development, discovery and publication of decision applications. Applications can be submitted from a user to a data warehouse in association with a data feed. One or more discovery properties are determined with regard to each application. The applications are made available to other client systems in association with the data feed. A relevant data feed and a relevant application can be identified based on satisfaction of a discovery request by the one or more determined discovery properties of the application. The application can be selected and downloaded to the user for evaluation and customization. The customized application can then be submitted to the data warehouse for publication with the other applications associated with the data feed.

Business application publication

The tracking of physical entities within a space such that computer-implemented computations (or ambient computing) may be performed on the features and/or activity of physical entities within that physical space. A space-time data structure corresponding to the physical space is set up. For at least one physical entity within that physical space, physical feature(s) of the physical entity is sensed at different instances in time. That sensed feature is stored in an entity data structure that is associated with the space-time data structure. This may be done for multiple physical entities within the location using multiple entity data structures that are each associated with the space-time data structure. A computing system may navigate such a space-time data structure to perform computation on physical entities, physical features and physical activity within that physical space.

Computer-aided tracking of physical entities

The subject disclosure is directed towards a web service that maintains a set of models used to generate plans, such as vacation plans, in which the set of models includes models that are authored by crowd contributors via the service. The models include rules, constraints and/or equations, and may be text based and declarative such that any author can edit an existing model or combination of existing models into a new model. Users can access the models to generate a plan according to user parameters, view a presentation of that plan, and interact to provide new parameters to the model and/or with objects in the plan to modify the plan and view a presentation of the modified plan.

Addition of plan-generation models and expertise by crowd contributors

Growing community of inventors

Kirkland, WA, United States of America

Vijay Mital