Deep command search within and across applications

Patent metadata

Field	Value
Publication number	US-11347756-B2
Application number	US-201916551265-A
Country	US
Kind code	B2
Filing date	Aug 26, 2019
Priority date	Aug 26, 2019
Publication date	May 31, 2022
Grant date	May 31, 2022

Abstract

Described herein are mechanisms to allow users to access functionality of applications in a suite of applications. In a first aspect, when a query relating to functionality of an application is received from a user, an index containing both top-level and sub-level functionality is searched. Results are ranked using a trained machine learning model using both context describing user interactions and the search results. A subset of the ranked results are presented to the user as options. In a second aspect the index can comprise entries describing functionality from other applications so that results presented to the user can include cross-application functionality. In a third aspect, the index can be searched using the context prior to receiving a query and adjusting the user interface based on the results. In a fourth aspect, the system can recommend other applications and/or devices that are better suited to a user's intent.

First claim

What is claimed is:

1. A method for command search, comprising:
receiving, by way of a productivity application that belongs to a suite of productivity applications, a query from a user, the query representing a request about functionality of the application;
searching an index using the query, the index indexes information about both top level and sub-top level functionality of the application, the top level and sub-top level functionality is accessible to the user within the application;
receiving results in response to the search;
accessing context for the user with respect to the application, the context for the user comprising previous interactions of the user with the application;
presenting the context for the user and the results to a trained machine learning model, the trained machine learning model having been previously trained using a combination of the context for the user and actions taken by the user subsequent to the previous interactions of the user with the application;
receiving, from the trained machine learning model, a ranking for each result in the subset;
selecting a result from the ranked results; and
presenting the selected result to the user.

2. The method of claim 1 further comprising:
identifying a top-level command accessible by the user through a user interface of the application;
identifying a sub-level command of the top-level command, the sub-level command accessible by the user through a second user interface;
creating an index entry in the index for the top-level command, the index entry comprising information about the top-level command; and
creating a second index entry in the index for the sub-level command, the second index entry comprising information about the sub-level command.

3. The method of claim 2 wherein the second index entry further comprises one or more parameters associated with the sub-level command.

4. The method of claim 1 wherein the trained machine learning model has been previously trained additionally using aggregate context for a plurality of users.

5. The method of claim 1 wherein the trained machine learning model has been previously trained using aggregate context for a plurality of users from a common tenancy.

6. The method of claim 1 further comprising:
monitoring the user's interaction with the selected result;
capturing the user's interaction and the context with respect to the user's interaction;
using the captured user's interaction and context to adjust parameters in the trained machine learning model to further train the trained machine learning model.

7. The method of claim 1 further comprising selecting the trained machine learning model from among a plurality of trained machine learning models.

8. The method of claim 1 wherein the trained machine learning model resides on a service and wherein presenting the context for the user and the results comprises sending the context and the results to the service.

9. The method of claim 1 wherein the trained machine learning model resides on the machine executing the method of claim 1 .

10. The method of claim 1 wherein the index further comprises functionality accessible in an external application and wherein the plurality of results received in response to the search comprise results related to the application and results related to the external application.

11. A system comprising a processor and memory, the memory storing computer executable instructions that, when executed by the processor, cause the processor to perform operations comprising:
receiving, by way of a productivity application that belongs to a suite of productivity applications, a query from a user, the query representing a request about functionality of the application;
searching an index using the query, the index indexes information about both top level and sub-top level functionality of the application, the top level and sub-level functionality is accessible to the user within the application;
receiving results in response to the search;
accessing context for the user with respect to the application, the context for the user comprising previous interactions of the user with the application;
presenting the context for the user and the results to a trained machine learning model, the trained machine learning model having been previously trained using a combination of the context for the user and actions taken by the user subsequent to the previous interactions of the user with the application;
receiving, from the trained machine learning model, a ranking for each result in the subset;
selecting a result from the ranked results; and
presenting the selected result to the user.

12. The system of claim 11 wherein the operations further comprise:
identifying a top-level command accessible by the user through a user interface of the application;
identifying a sub-level command of the top-level command, the sub-level command accessible by the user through a second user interface;
creating an index entry in the index for the top-level command, the index entry comprising information about the top-level command; and
creating a second index entry in the index for the sub-level command, the second index entry comprising information about the sub-level command.

13. The system of claim 12 wherein the second index entry further comprises one or more parameters associated with the sub-level command.

14. The system of claim 11 wherein the operations further comprise:
receiving a registration request from a second application comprising information regarding functionality of the second application; and
creating at least one entry in the index related to the information received in the registration request.

15. The system of claim 11 wherein the operations further comprise:
sending the query to a second application;
receiving at least one search result from the second application; and
presenting the at least one search result to the user with the selected result.

16. The system of claim 11 wherein the operations further comprise:
monitoring the user's interaction with the selected result;
capturing the user's interaction and the context with respect to the user's interaction; and
using the captured user's interaction and context to adjust parameters in the trained machine learning model to further train the trained machine learning model.

17. The system of claim 11 wherein the operations further comprise:
prior to receiving the query:
accessing the context;
searching the index using information from the context;
ranking results received responsive to searching the index using information from the context;
selecting a subset of the ranked results; and
modifying at least one user interface to present the selected subset.

18. A computer storage medium comprising executable instructions that, when executed by a processor of a machine, cause the machine to perform operations comprising:
receiving, by way of a productivity application that belongs to a suite of productivity applications, a query from a user, the query representing a request about functionality of the application;
searching an index using the query, the index indexes information about both top level and sub-top level functionality, the top level and sub-top level functionality is accessible to the user within the application;
receiving results in response to the search;
accessing context for the user with respect to the application, the context for the user comprising previous interactions of the user with the application;
presenting the contex

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Kleiner Madeline Schuster
• Freier Nathaniel George
• Kohlmeier Bernhard
• Agashe Amod Anil
• Shah Jignesh
• Religa Tomasz Lukasz

Classifications

• G06N3/045

  Combinations of networks · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/08

  Learning methods · CPC title

• G06F16/22

  Indexing; Data structures therefor; Storage structures · CPC title