The Aro Corporation

Microsoft Technology Licensing, LLC

Vulcan Technologies LLC

Kiha Software Inc.

Michael Perkowitz

Kevin Francis Eustice

Andrew Francis Hickl

Jonathan D Lazarus

Kenji Kawai

Alan Linchuan Liu

Meliha Yetisgen-Yildiz

Paul G Allen

Patrick James Ferrel

John Eugene Ball

Oliver Crandall Johnson

Erin Mounts

Ian G Clifton

Steven M Reed

Ned Dykstra Hayes

Nolan Lawson

Michael Zimmerman

Brian DeWeese

Kenneth Dwight Krossa

Embodiments create and label contextual slices from observation data and aggregate slices into a hierarchical storyline for a user. A context is a (possibly partial) specification of what a user was doing in the dimensions of time, place, and activity. A storyline is composed of a time-ordered sequence of contexts that partition a given span of time that are arranged in groups at one or more hierarchical levels. A storyline is created through a process of data collection, slicing, labeling, and aggregating. Raw context data can be collected from a variety of observation sources with various error characteristics. Slicing refines the raw context data into a consistent storyline composed of a sequence of contexts representing homogeneous time intervals. Labeling adds more specific and semantically meaningful data (e.g., geography, venue, activity) to the slices. Aggregation identifies groups of slices that correspond to a single semantic concept.

Automatically creating a hierarchical storyline from mobile device data

A user interface for an electronic calendar represents different locations or different users or different user calendars in different portions of the display. Calendar entries can be associated with one or more locations, one or more users, and with one or more user calendars. The different locations may reside in different time zones and a timeline for each time zone is displayed. The position of the calendar entry provides a visual identifier of the timeline with which the event is associated. Travel time to and from events in the calendar are calculated for calendared events and shown adjacent to the beginning and end of the event. A user's future location at a point in time is inferred from patterns in the user's locations and by analyzing the user's calendared events and correspondence in order to calculate travel time to calendared events.

Calendaring location-based events and associated travel

When a mobile device is known to be at a venue, readings from one or more sensors of the device are used to generate a sensor fingerprint of the venue and/or a venue category corresponding to the venue. The sensor fingerprint indicates typical sensor readings for the one or more physical sensors for devices that are located at the venue and/or venues of the same category. Sensor fingerprints can also be generated for other circumstances associated with the mobile device, such as the activity being undertaken by the corresponding user and a type of event currently occurring at the venue. Fingerprints can also be generated for combinations of circumstances, such as a particular activity at a certain category of venue. When location data cannot distinguish at which venue a mobile device is located, sensor fingerprints are compared to sensor readings received from the device to resolve this ambiguity.

Venue identification using sensor fingerprints

A mobile device includes a plurality of sensors, each with one or more costs associated with it. In order to reduce the cost associated with using the sensors to infer information about the device's current context, a sensor manager first collects readings from relatively low-cost sensors, and attempts to infer the device's context based on these readings. If the context cannot be unambiguously determined (within an acceptable degree of tolerance) using the low-cost sensor readings, the sensor manager activates one or more higher-cost sensors to identify the current context. In some instances, if the higher-cost sensor is still not adequate to determine the context, one or more even higher-cost sensors are activated. The weighting of the various costs associated with the sensors can be adjusted based on previous and/or predicted contexts.

Accurate mobile context detection at low sensor cost

A social network model, based on data relevant to a user, is used for semantic processing to enable improved entity recognition among text accessed by the user. An entity extraction module of the server, with reference to a general training corpus, general gazetteers, user-specific gazetteers, and entity models, parses text to identify entities. The entities may be, for example, people, organizations, or locations. A social network module of the server builds the social network model implicit in the data accessed by the user. The social network model includes the relationships between entities and an indication of the strength of each relationship. The social network module is also used to disambiguate names and unify entities based on the social network model.

Social network model for semantic processing

A user's context history is analyzed to identify transitions between contexts therein. The identified transitions are used to build a routine model for the user. The routine model includes transition rules indicating a source context, a destination context, and, optionally, a probability that the user will transition from the source context to the destination context, based on the user's historical behavior. A customized recommendation agent for the user is built using the routine model. The customized recommendation agent selects recommendations from a corpus to present to the user, based on the routine model and the user's current or predicted future context.

Recommendation agent using a routine model determined from mobile device data

A user query including one or more references to objects of various types and/or text keywords is received from a user and processed into sub-queries. Information sources execute the sub-queries and returns search results matching text keywords and/or relating to objects referenced in the user queries. The search results are organized based on their object types and relationships, and displayed in a manner exposing their relationships to the user.

Object-based relationship search using a plurality of sub-queries

A user's context history is analyzed to build a personality model describing the user's personality and interests. The personality model includes a plurality of metrics indicating the user's position on a plurality of personality dimensions, such as desire for novelty, tendency for extravagance, willingness to travel, love of the outdoors, preference for physical activity, and desire for solitude. A customized recommendation agent is then built based on the personality model, which selects a recommendation from a corpus to present to the user based on an affinity between the user's personality and the selected recommendation.

Recommendation agent using a personality model determined from mobile device data

A user's context history is used to help select contextual information to provide to the user. Context data describing the user's current context is received and a plurality of information items corresponding to the user's current context are identified from a contextual information corpus. A personalized user behavior model for the user is applied to determine the likelihood that each of the identified information items will be of value to the user. One or more of the information items are selected based on the corresponding likelihoods and the selected information items are provided for presentation to the user.

Contextual information provider

An entity recognition model is trained to use gazetteers to recognize entities referenced in documents. When a document associated with a user is received for recognizing entities referenced therein, a gazetteer is selected based on user contexts associated with the user. The document is analyzed using the gazetteer according to the entity recognition model to recognize an entity referenced therein. Additional facts associated with the recognized entities and actions applicable to the recognized entities are provided to enable the user to perform the applicable actions against the recognized entity.

Dynamic gazetteers for entity recognition and fact association

A system, method, and computer program product dynamically prioritizes electronic messages. An electronic message having one or more properties is received. These message properties can include, a particular sender or body text. Information describing past activity of a recipient user of the electronic message is accessed. A priority is determined for the electronic message, where the determining is based at least in part on a comparison of a property of the electronic message with the accessed information. The priority determination may include detecting the presence of a request in the electronic message, determining the social weight of the sender of the electronic message, determining the temporal urgency of the electronic message, or determining the relevance of the electronic message, for example. An indication of the priority of the message is presented to the recipient user.

Electronic message prioritization

The present invention is directed to sharing and organizing world wide web pages using a number of distinctive characters, each having an interest. In a preferred embodiment, each distinctive character is initialized by associating with it one or more web pages that collectively characterize the interest of the character. After initialization, each web page viewed by the user is assessed for each of the distinctive characters. The assessment process involves, for a single distinctive character, determining whether the web page viewed by the user exhibits at least a threshold level of similarity to web pages already associated with the distinctive character. The assessment process further involves associating the viewed web page with the distinctive character if the viewed web page exhibits at least a threshold level of similarity to the web pages already associated with the distinctive character. As a result, after the user has viewed a number of web pages, the viewed web pages that characterize each distinctive character's interest are associated with that distinctive character.

Growing community of inventors

Seattle, WA, United States of America

Michael Perkowitz