Personalized contextual suggestion engine

Abstract

Aspects of the technology described herein provide a more efficient user interface by providing suggestions that are tailored to a specific user’s interests. The suggestions may be provided by a personal assistant or some other application running on a user’s computing device. A goal of the technology described herein is to provide relevant suggestions when the user can and actually wants to use them. The suggestions are designed to provide information or services the user wants to use.

Claims

CLIAMS:

1. A computing system comprising:a processor; andcomputer storage memory having computer-executable instructions stored thereon which, when executed by the processor, implement a method of generating user-specific contextual association rules, the method comprising:(1) analyzing signal information associated with a computing device to determine an event of a first type has occurred, the computing device being associated with a user;(2) building a data vector for the event by assigning a value to a plurality of components of the data vector using information derived from the signal information; (3) generating an association rule for the first type of event using the data vector as an input to a machine learning algorithm that generates the association rule as an output, the association rule assigning a probability that the user is interested in a specific class of contextual suggestion given a specific user context; and(4) storing the association rule.

2. The system of claim 1, wherein the event is the user listening to music and the plurality of components of the data vector comprise a user location, a time, and a genre of the music.

3. The system of claim 1, wherein the event is the user watching a movie and the plurality of components of the data vector comprise a user location, a time, and a source of the movie.

4. The system of claim 1, wherein the specific type of contextual suggestion is show times for one or more movies showing at a movie theater within a threshold distance of a location indicated by the specific user context.

5. The system of claim 1, wherein the event is the user eating at a restaurant and the plurality of components of the data vector comprise a user location, a time, and a food type served by the restaurant.

6. The system of claim 1, wherein the method further comprises enhancing the signal information by identifying an entity within the signal information and retrieving supplemental information about the entity from a knowledge base.

7. The system of claim 6, wherein the knowledge base comprises information about the user that is collected, at least in part, by a personal assistant application running on the computing device.

8. A method of generating user-specific contextual association rules, the method comprising:receiving signal information describing an event from a computing device associated with a user;classifying the event as non-routine;building a data vector for the event by assigning a value to a plurality of components of the data vector using information derived from the signal information, wherein a component of the data vector indicates that the event is non-routine; generating an association rule for a first type of event using the data vector as an input to a machine learning algorithm that generates the association rule as an output, the association rule assigning a probability that the user is interested in a contextual suggestion given a specific non-routine user context; and9. The method of claim 8, wherein the method further comprises:identifying an entity within the signal information; andretrieving supplemental information about the entity from a knowledge base, and wherein at least one of the plurality of components of the data vector is associated with a value derived from the supplemental information.

10. The method of claim 9, wherein the knowledge base comprises information about the user.

11. The method of claim 10, wherein at least a portion of the information is collected by a personal assistant running on the computing device.

12. The method of claim 8, wherein said classifying the event as non-routine is performed by an automated classifier.

13. The method of claim 8, wherein the data vector describes a dining event and includes a restaurant name component, a restaurant type component, and a time component.

14. The method of claim 8, wherein the method further comprises:receiving additional signal information from a device associated with the user;determining a present context of the user from the additional signal information;determining that the present context is the specific non-routine user context; andtriggering the contextual suggestion.

15. The method of claim 8, wherein the computing device is one of a smartphone or a personal computer.

16. A method of generating user-specific contextual association rules comprising:receiving signal information from one or more devices associated with a user;enhancing the signal information by identifying an entity within the signal information and retrieving supplemental information about the entity from a knowledge base;building a data vector by assigning a value to each of a plurality of components of the data vector, wherein the value is from the signal information or the supplemental information;generating an association rule for the user using the data vector as an input to a machine learning algorithm that generates the association rule as an output, the association rule assigning a probability that the user is interested in a contextual suggestion given a specific user context; and17. The method of claim 16, wherein the method further comprises:determining that the present context matches the specific user context in the association rule; and18. The method of claim 16, wherein the knowledge base comprises information about the user.

19. The method of claim 16, wherein the data vector comprises a component indicating that an event described by the data vector is a routine event.

20. The method of claim 16, wherein the knowledge base describes a type of restaurant. As Attached