Systems and methods for proactively providing recommendations to a user of a computing device

Patent metadata

Field	Value
Publication number	US-10922094-B2
Application number	US-201514866795-A
Country	US
Kind code	B2
Filing date	Sep 25, 2015
Priority date	Jun 5, 2015
Publication date	Feb 16, 2021
Grant date	Feb 16, 2021

Abstract

The embodiments described herein set forth techniques for identifying when a user activates a search application on his or her mobile computing device, and presenting, prior to receiving an input of search parameters from the user, a prediction of one or more applications that the user may be interested in accessing. According to some embodiments, the search application can be configured to interface with an “application prediction engine” each time the search application is activated and query the application prediction engine for a prediction of one or more applications that the user may be interested in accessing. In turn, the application prediction engine can analyze information associated with the applications installed on the mobile computing device to produce the prediction. Using the prediction, the search application can display the predicted one or more applications within a user interface of the search application for selection by the user.

First claim

What is claimed is:

1. A method for proactively providing predictions to a user of a mobile device, the method comprising, at a prediction center executing on the mobile device:
for each prediction engine of a plurality of prediction engines executing locally on the mobile device:
receiving, from the prediction engine, a respective request for the prediction engine to function as an expert on at least one prediction category of a plurality of prediction categories, and
updating a configuration of the prediction center to reflect a registration of the prediction engine as an expert on the at least one prediction category;

receiving, from a search application executing on the mobile device, a request for an optimized list of applications that the user is likely to activate, wherein the request corresponds to a particular prediction category, and the particular prediction category relates to applications that the user is likely to activate;
identifying, among the plurality of prediction engines, at least two prediction engines that are registered as experts on the particular prediction category;
issuing, to each prediction engine of the at least two prediction engines, requests for respective lists of applications that the user is likely to activate;
receiving the respective lists from the at least two prediction engines;
processing the respective lists of applications to generate the optimized list of applications; and
providing the optimized list to the search application.

2. The method of claim 1 , wherein processing the respective lists of applications to establish the optimized list of applications comprises:
removing redundant information;
applying respective weights to each of the respective lists, wherein each respective weight is based on historical accuracy metrics associated with the prediction engine that provides the respective list; and/or
sorting one or more of the respective lists or the optimized list in accordance with scores included therein, wherein each score corresponds to a likelihood that a particular application will be activated by the user.

3. The method of claim 2 , wherein applying the respective weights comprises:
adjusting the scores in accordance with the respective weights.

4. The method of claim 2 , wherein each prediction engine of the at least two prediction engines establishes respective scores for each application by performing at least one function on at least one data signal that corresponds to the application.

5. The method of claim 4 , wherein the at least one data signal that corresponds to an application is selected from one or more of:
application installation timestamps,
application activation timestamps,
application activation totals,
application usage metrics,
positions of application icons within a main user interface of the mobile device,
search parameters recently provided by the user, and
gathered feedback that indicates whether previous predictions were accurate.

6. The method of claim 5 , wherein a position of an application icon for a given application within the main user interface of the mobile device indicates:
a page number of the main user interface in which the application icon is included, or
an indication of whether the application icon is included in a folder within the main user interface.

7. The method of claim 5 , further comprising, subsequent to providing the optimized list to the search application:
receiving feedback from the search application, wherein the feedback indicates a behavior of the user subsequent to viewing the optimized list of applications in the search application; and
updating the gathered feedback to reflect the feedback received from the search application.

8. At least one non-transitory computer readable storage medium configured to store instructions that, when executed by at least one processor included in a mobile device, cause the mobile device to proactively provide predictions to a user of the mobile device, by carrying out steps that include:
for each prediction engine of a plurality of prediction engines executing locally on the mobile device:
receiving, from the prediction engine, a respective request for the prediction engine to function as an expert on at least one prediction category of a plurality of prediction categories, and
updating a configuration of the prediction center to reflect a registration of the prediction engine as an expert on the at least one prediction category;

receiving, from a search application executing on the mobile device, a request for an optimized list of applications that the user is likely to activate, wherein the request corresponds to a particular prediction category, and the particular prediction category relates to applications that the user is likely to activate;
identifying, among the plurality of prediction engines, at least two prediction engines that are registered as experts on the particular prediction category;
issuing, to each prediction engine of the at least two prediction engines, requests for respective lists of applications that the user is likely to activate;
receiving the respective lists from the at least two prediction engines;
processing the respective lists of applications to generate the optimized list of applications; and
providing the optimized list to the search application.

9. The at least one non-transitory computer readable storage medium of claim 8 , wherein processing the respective lists of applications to establish the optimized list of applications comprises:
removing redundant information;
applying respective weights to each of the respective lists, wherein each respective weight is based on historical accuracy metrics associated with the prediction engine that provides the respective list; and/or
sorting one or more of the respective lists or the optimized list in accordance with scores included therein, wherein each score corresponds to a likelihood that a particular application will be activated by the user.

10. The at least one non-transitory computer readable storage medium of claim 9 , wherein applying the respective weights comprises:
adjusting the scores in accordance with the respective weights.

11. The at least one non-transitory computer readable storage medium of claim 9 , wherein each prediction engine of the at least two prediction engines establishes respective scores for each application by performing at least one function on at least one data signal that corresponds to the application.

12. The at least one non-transitory computer readable storage medium of claim 11 , wherein the at least one data signal that corresponds to an application is selected from one or more of:
application installation timestamps,
application activation timestamps,
application activation totals,
application usage metrics,
positions of application icons within a main user interface of the mobile device,
search parameters recently provided by the user, and
gathered feedback that indicates whether previous predictions were accurate.

13. The at least one non-transitory computer readable storage medium of claim 12 , wherein a position of an application icon for a given application within the main user interface of the mobile device indicates:
a page number of the main user interface in which the application icon is included, or
an indication of whether the application icon is included in a folder within the main user interface.

14. The at least one non-transitory computer readable storage medium of claim 12 , wherein the steps further include, subsequent to providing the optimized list to the search application:
r

Assignees

• Apple Inc

Inventors

• Morris Colin
• Scott Peter J
• Federighi Craig M
• Gross Daniel C

Classifications

• H04M1/72472

  wherein the items are sorted according to specific criteria, e.g. frequency of use · CPC title

• H04M1/72454

  according to context-related or environment-related conditions · CPC title

• G06F16/903

  Querying (for retrieval from the web G06F16/953) · CPC title

• G06F9/453 Primary

  Help systems · CPC title

• G06F9/4451 Primary

  User profiles; Roaming · CPC title