Behavior prediction

• US 10,540,599 B2
• Filed: 04/07/2016
• Issued: 01/21/2020
• Est. Priority Date: 04/07/2016
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Chander, Ajay, Malur Srinivasan, Ramya
• Primary Examiner(s): Al-Hashemi, Sana A
• Application Number:
  US15/093,662
• Publication Number:
  US 20170293853A1
• Time in Patent Office: 1,384 Days
• Field of Search: 707607, 707609, 707687, 707705, 707718, 707790, 707813, 707821
• CPC Class Codes: G06F 3/0481 based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A method may include extracting feature information from a user history including multiple goal messages. Each of the multiple goal messages may include visual media and text, and may be associated with a goal indicated as completed. Extracting the feature information from the user history may include performing feature detection on the visual media of the multiple goal messages. The feature information from the user history may be analyzed for correlations between features and the completion indications of the associated goals. Feature information may be extracted from a generic goal message. A probability of a user indicating completion of a goal associated with the generic goal message may be determined based at least in part on the feature information from the generic goal message and the correlations between the features included in the multiple goal messages and the indications of the associated goals having been completed.

Claims

1. A method comprising:

obtaining a plurality of previous goal messages from a user history of a user, each respective previous goal message of the plurality of previous goal messages including previous visual presentation features that relate to how suggestion of performance of a respective action related to accomplishing a corresponding goal of the respective previous goal message is visually presented to the user for completion by the user, the visual presentation features including;

text describing the respective action and visual media related to the respective action described by the corresponding text of the respective goal message;

obtaining, from the user history, a plurality of indications in which each respective indication of the plurality of indications is with respect to a corresponding one respective previous goal message of the plurality of previous goal messages, the respective indication indicating whether the respective action of its corresponding respective previous goal message was performed by the user;

extracting historical feature information related to the visual presentation features of the plurality of previous goal messages and completion of the respective actions of the previous goal messages, the extracting of the historical feature information being based on one or more of;

the text of the plurality of previous goal messages, the visual media of the plurality of previous goal messages, and the plurality of indications;

analyzing the historical feature information and the plurality of indications to determine correlations between how the previous visual presentation features included in the plurality of previous goal messages relate to the indications of the performance of the respective actions of the plurality of previous goal messages;

extracting feature information from a generic goal message, the feature information relating to visual presentation features of the generic goal message that relate to how a suggested action of the generic goal message is visually presented to the user for completion by the user;

determining, based on the historical feature information and the feature information from the generic goal message, a visual relationship between how the previous visual presentation features relate to the visual presentation features of the generic goal message with respect to visual presentation of their respective actions; and

determining a probability of the user indicating completion of the suggested action associated with the generic goal message based at least in part on the visual relationship and the determined correlations.

2. The method of claim 1, further comprising:

in response to the probability of the user indicating completion of the suggested action associated with the generic goal message being below a probability threshold, generating a personalized goal message based on the generic goal message including one or more revised features; and

delivering the personalized goal message to a device associated with the user.

3. The method of claim 2, wherein the one or more revised features of the personalized goal message is associated with an increase in the probability of the user indicating completion of the suggested action associated with the personalized goal message relative to a goal associated with the generic goal message.

4. The method of claim 1, further comprising, in response to the probability of the user indicating completion of the suggested action associated with the generic goal message being at or above a probability threshold, delivering a personalized goal message equivalent to the generic goal message to a device associated with the user.

5. The method of claim 1, wherein:

the extracting of historical feature information includes detecting human faces; and

the determined correlations are based on expressions of the detected human faces.

6. The method of claim 1, wherein:

the extracting of historical feature information includes at least one of determining emotions of the previous visual presentation features and determining emotions of detected human faces; and

the determined correlations are based on at least one of the determined emotions of the visual media and the determined emotions of the detected human faces.

7. The method of claim 1, wherein the extracting of the historical feature information includes identifying positive words or phrases included in the text via text analysis.

8. The method of claim 1, wherein the extracting of the historical feature information includes identifying one or more user features of the user, the one or more user features including one or more of:

a total number of goals indicated as completed by the user, and a theme preference of the user.

9. A non-transitory computer-readable medium having encoded therein programing code executable by a processor to perform operations comprising:

obtaining a plurality of previous goal messages from a user history of a user, each respective previous goal message of the plurality of previous goal messages including previous visual presentation features that relate to how suggestion of performance of a respective action related to accomplishing a corresponding goal of the respective previous goal message is visually presented to the user for completion by the user, the visual presentation features including;

text describing the respective action and visual media related to the respective action described by the corresponding text of the respective goal message;

obtaining, from the user history, a plurality of indications in which each respective indication of the plurality of indications is with respect to a corresponding one respective previous goal message of the plurality of previous goal messages, the respective indication indicating whether the respective action of its corresponding respective previous goal message was performed by the user;

extracting historical feature information related to the visual presentation features of the plurality of previous goal messages and completion of the respective actions of the previous goal messages, the extracting of the historical feature information being based on one or more of;

the text of the plurality of previous goal messages, the visual media of the plurality of previous goal messages, and the plurality of indications;

analyzing the historical feature information and the plurality of indications to determine correlations between how the previous visual presentation features included in the plurality of previous goal messages relate to the indications of the performance of the respective actions of the plurality of previous goal messages;

extracting feature information from a generic goal message, the feature information relating to visual presentation features of the generic goal message that relate to how a suggested action of the generic goal message is visually presented to the user for completion by the user;

determining, based on the historical feature information and the feature information from the generic goal message, a visual relationship between how the previous visual presentation features relate to the visual presentation features of the generic goal message with respect to visual presentation of their respective actions; and

determining a probability of the user indicating completion of the suggested action associated with the generic goal message based at least in part on the visual relationship and the determined correlations.

10. The non-transitory computer-readable medium of claim 9, the operations further comprising:

in response to the probability of the user indicating completion of the suggested action associated with the generic goal message being below a probability threshold, generating a personalized goal message based on the generic goal message including one or more revised features; and

delivering the personalized goal message to a device associated with the user.

11. The non-transitory computer-readable medium of claim 10, wherein the one or more revised features of the personalized goal message is associated with an increase in the probability of the user indicating completion of the suggested action associated with the personalized goal message relative to a goal associated with the generic goal message.

12. The non-transitory computer-readable medium of claim 9, the operations further comprising, in response to the probability of the user indicating completion of the suggested action associated with the generic goal message being at or above a probability threshold, delivering a personalized goal message equivalent to the generic goal message to a device associated with the user.

13. The non-transitory computer-readable medium of claim 9, wherein:

the extracting of historical feature information includes detecting human faces; and

the determined correlations are based on expressions of the detected human faces.

14. The non-transitory computer-readable medium of claim 9, wherein:

the extracting of historical feature information includes at least one of determining emotions of the previous visual presentation features and determining emotions of detected human faces; and

the determined correlations are based on at least one of the determined emotions of the visual media and the determined emotions of the detected human faces.

15. The non-transitory computer-readable medium of claim 9, wherein the extracting of the historical feature information includes identifying positive words or phrases included in the text via text analysis.

16. A system comprising:

one or more processors; and

one or more computer-readable media having encoded therein programing code executable by the one or more processors, that, in response to being executed by the one or more processors, causes the system to perform operations comprising;

obtaining a plurality of previous goal messages from a user history of a user, each respective previous goal message of the plurality of previous goal messages including previous visual presentation features that relate to how suggestion of performance of a respective action related to accomplishing a corresponding goal of the respective previous goal message is visually presented to the user for completion by the user, the visual presentation features including;

text describing the respective action and visual media related to the respective action described by the corresponding text of the respective goal message;

obtaining, from the user history, a plurality of indications in which each respective indication of the plurality of indications is with respect to a corresponding one respective previous goal message of the plurality of previous goal messages, the respective indication indicating whether the respective action of its corresponding respective previous goal message was performed by the user;

extracting historical feature information related to the visual presentation features of the plurality of previous goal messages and completion of the respective actions of the previous goal messages, the extracting of the historical feature information being based on one or more of;

the text of the plurality of previous goal messages, the visual media of the plurality of previous goal messages, and the plurality of indications;

analyzing the historical feature information and the plurality of indications to determine correlations between how the previous visual presentation features included in the plurality of previous goal messages relate to the indications of the performance of the respective actions of the plurality of previous goal messages;

extracting feature information from a generic goal message, the feature information relating to visual presentation features of the generic goal message that relate to how a suggested action of the generic goal message is visually presented to the user for completion by the user;

determining, based on the historical feature information and the feature information from the generic goal message, a visual relationship between how the previous visual presentation features relate to the visual presentation features of the generic goal message with respect to visual presentation of their respective actions; and

determining a probability of the user indicating completion of the suggested action associated with the generic goal message based at least in part on the visual relationship and the determined correlations.

17. The system of claim 16, wherein the operations further comprise:

in response to the probability of the user indicating completion of the suggested action associated with the generic goal message being below a probability threshold, generating a personalized message based on the generic goal message including one or more revised features and delivering the personalized message to a device associated with the user; and

in response to the probability of the user indicating completion of the suggested action associated with the generic goal message being at or above the probability threshold, delivering a personalized message equivalent to the generic goal message to the device associated with the user.

18. The system of claim 16, wherein the extracting of historical feature information includes at least one of determining emotions of the visual media and determining emotions of detected human faces.

19. The system of claim 16, wherein the extracting of the historical feature information includes identifying positive words or phrases included in the text via text analysis.

20. The system of claim 16, wherein the extracting of the historical feature information includes identifying one or more user features of the user, the one or more user features including one or more of:

a total number of goals indicated as completed by the user, and a theme preference of the user.