Granted Patent B2

US 9,654,437 · App. 14/225,241 · Granted May 16, 2017

Inventors: Jianqiang Shen (Santa Clara, CA); Masafumi Suzuki (Mountain View, CA); Oliver Brdiczka (Mountain View, CA)

Assignee: PALO ALTO RESEARCH CENTER INCORPORATED

H04L51/26G10L15/063G10L15/08H04L51/12H04L51/24H04L67/22H04L67/322H04L51/063

Abstract

Claims (80)

1. A computer-implemented method for selectively displaying messages to a user, the method comprising:

determining a priority of a message based on:

a sender of the message;

whether the message requests a response, wherein determining the priority of the message based on whether the message requests a response involves training a speech predictor to predict speech acts from the message;

a duration of time within which the user can read the message while driving a vehicle; and

one or more rules;

responsive to determining that the user is not an intended recipient of the message, decreasing the priority of the message; and

precluding display of one or more messages while the user is driving the vehicle based on the respective priority of the messages.

2. The method of claim 1 , wherein determining the priority of the message further comprises using a weighted score for each of the respective measures of the priority of the message based on predetermined multipliers.

3. The method of claim 1 , wherein determining the priority of the message based on the sender of the message further comprises calculating a perceived priority of a communication pattern of the user.

4. The method of claim 1 , wherein the speech predictor is trained from a collection of messages using machine learning methods, and wherein determining the priority of the message based on whether the message requests a response further comprises:

scoring the message using the speech predictor.

5. The method of claim 1 , wherein determining the priority of the message based on the duration of time within which the user can read the message while driving the vehicle further comprises counting the number of words in the message.

6. The method of claim 1 , wherein determining the priority of the message based on the one or more rules further comprises determining an overall score for the rules, wherein each rule contains a condition and a predetermined score corresponding to the condition, and wherein the conditions of the rules comprise one or more of:

whether a particular sender is associated with the message;

whether a particular word appears within the message; and

whether a sentence contains two particular words in a particular order; and

wherein if the conditions of the rules are present, the method further comprises determining the overall score for the rules by summing up the predetermined scores corresponding to the conditions of the rules.

7. The method of claim 1 , further comprising:

determining whether the user is an intended recipient of the message.

8. The method of claim 1 , further comprising:

partitioning the message into sentences; and

determining the priority of each sentence.

9. The method of claim 8 , further comprising displaying one or more messages based on the respective priority of the messages by one or more of:

displaying the sentences whose priority is greater than a predetermined threshold; and

displaying all of the sentences, regardless of their priority, wherein the sentences with a priority greater than a predetermined threshold are displayed in color.

10. The method of claim 8 , further comprising displaying one or more messages based on the respective priority of the messages by summarizing the messages to display to the user, which comprises one or more of:

sorting the sentences based on their priority;

determining a total character capacity of a display;

determining a character count of each sorted sentence;

selecting the sentences with the highest priority, wherein the summation of the character count of the selected sentences is equal to or less than the total character capacity of the display;

ordering the selected sentences in the order in which they appear in the message; and

displaying the ordered sentences on the display of the user.

11. A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for selectively displaying messages to a user, the method comprising:

determining a priority of a message based on:

a sender of the message;

whether the message requests a response, wherein determining the priority of the message based on whether the message requests a response involves training a speech predictor to predict speech acts from the message;

a duration of time within which the user can read the message while driving a vehicle; and

one or more rules,

wherein determining the priority of the message comprises combining measures of the priority determined based on the sender of the message, whether the message requests a response, the duration of time within which to read the message, and the one or more rules, and

wherein determining the priority of the message further comprises using a decision tree with branches which check each of the respective measures of the priority of the message against conditions based on a predetermined threshold;

responsive to determining that the user is not an intended recipient of the message, decreasing the priority of the message; and

precluding display of one or more messages while the user is driving the vehicle based on the respective priority of the messages.

12. The storage medium of claim 11 , wherein determining the priority of the message further comprises using a weighted score for each of the respective measures of the priority of the message based on predetermined multipliers.

13. The storage medium of claim 11 , wherein determining the priority of the message based on the sender of the message further comprises calculating a perceived priority of a communication pattern of the user.

14. The storage medium of claim 11 , wherein the speech predictor is trained from a collection of messages using machine learning methods, and wherein determining the priority of the message based on whether the message requests a response further comprises:

scoring the message using the speech predictor.

15. The storage medium of claim 11 , wherein determining the priority of the message based on the duration of time within which the user can read the message while driving the vehicle further comprises counting the number of words in the message.

16. The storage medium of claim 11 , wherein determining the priority of the message based on the one or more rules further comprises determining an overall score for the rules, wherein each rule contains a condition and a predetermined score corresponding to the condition, and wherein the conditions of the rules comprise one or more of:

whether a particular sender is associated with the message;

whether a particular word appears within the message; and

whether a sentence contains two particular words in a particular order; and

wherein if the conditions of the rules are present, the method further comprises determining the overall score for the rules by summing up the predetermined scores corresponding to the conditions of the rules.

17. The storage medium of claim 11 , further comprising:

determining whether the user is an intended recipient of the message.

18. The storage medium of claim 11 , further comprising:

partitioning the message into sentences; and

determining the priority of each sentence.

19. The storage medium of claim 18 , further comprising displaying one or more messages based on the respective priority of the messages by one or more of:

displaying the sentences whose priority is greater than a predetermined threshold; and

displaying all of the sentences, regardless of their priority, wherein the sentences with a priority greater than a predetermined threshold are displayed in color.

20. The storage medium of claim 18 , further comprising displaying one or more messages based on the respective priority of the messages by summarizing the messages to display to the user, which comprises one or more of:

sorting the sentences based on their priority;

determining a total character capacity of a display;

determining a character count of each sorted sentence;

selecting the sentences with the highest priority, wherein the summation of the character count of the selected sentences is equal to or less than the total character capacity of the display;

ordering the selected sentences in the order in which they appear in the message; and

displaying the ordered sentences on the display of the user.

21. A computer system to selectively display messages to a user, comprising:

a processor;

a storage device coupled to the processor and storing instructions that when executed by a computer cause the computer to perform a method, the method comprising:

determining a priority of a message based on:

a sender of the message;

whether the message requests a response, which involves training a speech predictor to predict speech acts from the message;

a duration of time within which the user can read the message while driving a vehicle; and

one or more rules,

wherein determining the priority of the message comprises combining measures of the priority determined based on the sender of the message, whether the message requests a response, the duration of time within which to read the message, and the one or more rules, and

wherein determining the priority of the message further comprises using a decision tree with branches which check each of the respective measures of the priority of the message against conditions based on a predetermined threshold;

responsive to determining that the user is not an intended recipient of the message, decreasing the priority of the message; and

precluding display of one or more messages while the user is driving the vehicle based on the respective priority of the messages.

Assignments (9)

Continuity (1)