Adaptive user interfaces

• US 8,775,332 B1
• Filed: 06/13/2013
• Issued: 07/08/2014
• Est. Priority Date: 06/13/2013
• Status: Active Grant

• Current Assignee: XANT Inc.
• Original Assignee: Insidesales.com
• Inventors: Morris, Richard Glenn, Zeng, Xinchuan, Elkington, David Randal
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): COUGHLAN, PETER D
• Application Number:
  US13/916,600
• Time in Patent Office: 390 Days
• Field of Search: 706/44, 706/11
• US Class Current: 706/11
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

According to various embodiments of the present invention, user performance and/or motivation for a computing system may be maximized by optimizing one or more target components of a user interface of the computing system. The target components may be aspects of the user interface that is perceived by the user. One or more input features and one or more output features may be identified, and data regarding these input and output features may be gathered. This data may be compared with the results generated by a set of candidate artificial intelligence algorithms to determine which of them provides the best fit with the data collected. Then, the selected artificial intelligence algorithm may be applied to the user interface to iteratively change the target components over time until the optimal settings for each user are discovered.

Claims

1. A computer-implemented method for optimizing a user interface of a computing system using artificial intelligence, comprising:

at a processor, identifying a first target component of the user interface, wherein the first target component is perceptible to a user via an output device;

at the processor, identifying at least one first output feature representing a measured indicator of at least one of performance and motivation of the user;

at the processor, identifying at least one first input feature comprising at least one of;

a first measured aspect descriptive of the user; and

a first measured aspect descriptive of the user'"'"'s interaction with the computing system;

at the processor, collecting data comprising a plurality of input measurements of the first input feature and a plurality of output measurements of the first output feature;

at the processor, selecting a first selected artificial intelligence algorithm to maximize the first output feature for the user, based on comparison of the plurality of input measurements and the plurality of output measurements with a plurality of candidate artificial intelligence algorithms; and

at the processor, applying the first selected artificial intelligence algorithm to maximize the first output feature for the user by changing the first target component.

2. The method of claim 1, further comprising:

at the processor, identifying a plurality of additional output features tracked by the computing system; and

at the processor, mathematically combining the plurality of additional output features with the first output feature to define a composite output feature;

wherein the data further comprises a plurality of output measurements of the composite output feature.

3. The method of claim 1, wherein the first output feature measures user activity tracked by the computing system.

4. The method of claim 1, wherein the first output feature measures user motivation in relation to activity tracked by the computing system.

5. The method of claim 1, further comprising:

at the processor, identifying a second output feature representing a measured indicator of at least one of performance and motivation of the user;

at the processor, identifying a second target component of the user interface, wherein the second target component is perceptible to the user via the output device;

at the processor, identifying a second input feature comprising at least one of;

a second measured aspect descriptive of the user; and

a second measured aspect descriptive of the user'"'"'s interaction with the computing system;

at the processor, selecting a second selected artificial intelligence algorithm to maximize the second output feature for the user, based on comparison of the plurality of input measurements and the plurality of output measurements with the plurality of candidate artificial intelligence algorithms; and

at the processor, applying the second selected artificial intelligence algorithm to maximize the second output feature for the user by changing the second target component;

wherein the data further comprises a plurality of input measurements of the second input feature and a plurality of output measurements of the second output feature.

6. The method of claim 1, wherein the first target component comprises a message delivered to the user by the computing system.

7. The method of claim 6, further comprising:

at the processor, identifying a plurality of candidate messages for the message; and

at the processor, ranking the candidate messages based at least in part on a frequency of use of language in each of the plurality of candidate messages outside the computing system;

wherein changing the first target component comprises changing the message to a selected message of the plurality of candidate messages.

8. The method of claim 1, wherein the output device comprises a display screen, and wherein the first target component comprises a content display policy that controls how the user interface displays information to the user on the display screen.

9. The method of claim 8, further comprising:

at the processor, quantifying possible changes to the content display policy based on user awareness of the possible changes; and

at the processor, determining a maximum change in the content display policy that will avoid distracting the user;

wherein changing the first target component comprises changing the content display policy by a variation no greater than the maximum change.

10. The method of claim 1, wherein the first target component comprises award criteria for issuance of an adaptive reward to the user, the method further comprising:

at the processor, determining that the user has accomplished a goal prior to issuance of the adaptive reward; and

notifying the user through the user interface that the adaptive reward has been issued.

11. The method of claim 10, wherein the award criteria comprises a plurality of user-specific award criteria, each of which applies to one of a plurality of users of the computing system.

12. The method of claim 1, further comprising:

at the processor, identifying a plurality of additional input features, each comprising at least one of;

a measured aspect descriptive of the user; and

a measured aspect descriptive of the user'"'"'s interaction with the computing system;

wherein the data further comprises a plurality of additional input measurements of each of the plurality of additional input features;

and wherein selecting the first selected artificial intelligence algorithm further comprises basing the selection on comparison of the plurality of additional input measurements and the plurality of output measurements with the plurality of candidate artificial intelligence algorithms.

13. The method of claim 1, wherein the first input feature comprises a personality type of the user.

14. The method of claim 1, wherein the first input feature comprises a demographic of the user.

15. The method of claim 1, wherein the first input feature comprises an action taken by the user relative to the user interface.

16. The method of claim 1, wherein the data further comprise a plurality of reward scores, each of which is associated with one of the plurality of input measurements.

17. The method of claim 16, wherein the data further comprises a plurality of quality scores, each of which is associated with one of the plurality of input measurements to indicate a quality of the associated input measurement relative to the other input measurements.

18. The method of claim 1, wherein collecting the data is carried out prior to at least one of identification of the at least one first output feature, identification of the first target component, and identification of the at least one first input feature.

19. The method of claim 1, wherein collecting the data is carried out after identification of the at least one first output feature, identification of the first target component, and identification of the at least one first input feature.

20. The method of claim 1, wherein selecting the first selected artificial intelligence algorithm comprises selecting at least one of:

an instance weight learning algorithm;

a genetic algorithm;

a single-layer perceptron;

a multi-layer perceptron;

a decision tree;

a support vector machine;

a naï

ve Bayes classifier;

a nearest neighbor-based algorithm;

a neural network;

a multiple output dependence algorithm;

a hierarchical based sequencing algorithm;

a multiple output relaxation algorithm;

a deep learning algorithm;

a derivative of an artificial intelligence algorithm; and

a combination of at least two artificial intelligence algorithms.

21. The method of claim 1, wherein applying the first selected artificial intelligence algorithm comprises:

at the processor, identifying a plurality of versions of the first target component; and

at the processor, iterating through the plurality of versions of the first target component to identify a selected version of the plurality of versions that yields optimal results for at least one of performance and motivation of the user.

22. The method of claim 1, wherein the first selected artificial intelligence algorithm comprises a machine learning algorithm.

23. A computer program product for optimizing a user interface of a computing system using artificial intelligence, comprising:

a non-transitory computer-readable storage medium; and

computer program code, encoded on the medium, configured to cause at least one processor to perform the steps of;

identifying a first target component of the user interface, wherein the first target component is perceptible to a user via an output device;

identifying at least one first output feature representing a measured indicator of at least one of performance and motivation of the user;

identifying at least one first input feature comprising at least one of;

a first measured aspect descriptive of the user; and

a first measured aspect descriptive of the user'"'"'s interaction with the computing system;

collecting data comprising a plurality of input measurements of the first input feature and a plurality of output measurements of the first output feature;

selecting a first selected artificial intelligence algorithm to maximize the first output feature for the user, based on comparison of the plurality of input measurements and the plurality of output measurements with a plurality of candidate artificial intelligence algorithms; and

applying the first selected artificial intelligence algorithm to maximize the first output feature for the user by changing the first target component.

24. The computer program product of claim 23, wherein the first target component comprises a message delivered to the user by the computing system, the computer program product further comprising computer program code, encoded on the medium, configured to cause at least one processor to perform the steps of:

identifying a plurality of candidate messages for the message; and

ranking the candidate messages based at least in part on a frequency of use of language in each of the plurality of candidate messages outside the computing system;

and wherein changing the first target component comprises changing the message to a selected message of the plurality of candidate messages.

25. The computer program product of claim 23, wherein the computer program code configured to cause at least one processor to apply the first selected artificial intelligence algorithm comprises computer program code configured to perform the steps of:

identifying a plurality of versions of the first target component; and

iterating through the plurality of versions of the first target component to identify a selected version of the plurality of versions that yields optimal results for at least one of performance and motivation of the user.

26. The computer program product of claim 23, wherein the first selected artificial intelligence algorithm comprises a machine learning algorithm.

27. A system for optimizing a user interface of a computing system using artificial intelligence, comprising:

an output device, configured to output at least one target component of the user interface;

in input device, configured to receive at least one input feature;

at least one processor, communicatively coupled to the output device and the input device, configured to;

identify a first target component of the user interface, wherein the first target component is perceptible to a user via the output device;

identify at least one first output feature representing a measured indicator of at least one of performance and motivation of the user;

identify at least one first input feature comprising at least one of;

a first measured aspect descriptive of the user; and

a first measured aspect descriptive of the user'"'"'s interaction with the computing system;

collect data comprising a plurality of input measurements of the first input feature and a plurality of output measurements of the first output feature;

select a first selected artificial intelligence algorithm to maximize the first output feature for the user, based on comparison of the plurality of input measurements and the plurality of output measurements with a plurality of candidate artificial intelligence algorithms; and

apply the first selected artificial intelligence algorithm to maximize the first output feature for the user by changing the first target component.

28. The system of claim 27, wherein the first target component comprises a message delivered to the user by the computing system, and wherein the at least one processor is further configured to:

identify a plurality of candidate messages for the message; and

rank the candidate messages based at least in part on a frequency of use of language in each of the plurality of candidate messages outside the computing system;

and wherein changing the first target component comprises changing the message to a selected message of the plurality of candidate messages.

29. The system of claim 27, wherein the at least one processor is configured to apply the first selected artificial intelligence algorithm by:

identifying a plurality of versions of the first target component; and

iterating through the plurality of versions of the first target component to identify a selected version of the plurality of versions that yields optimal results for at least one of performance and motivation of the user.

30. The system of claim 27, wherein the first selected artificial intelligence algorithm comprises a machine learning algorithm.