System and method for applying probability distribution models to dialog systems in the troubleshooting domain

• US 8,010,364 B2
• Filed: 10/31/2007
• Issued: 08/30/2011
• Est. Priority Date: 10/31/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Williams, Jason
• Primary Examiner(s): Han, Qi
• Application Number:
  US11/930,366
• Publication Number:
  US 20090112598A1
• Time in Patent Office: 1,399 Days
• Field of Search: 704/270, 704/270.1, 704/272, 704/275, 704/276, 704/231
• US Class Current: 704/270
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/5061 characterised by the interaction between service providers and their network customers, e.g. customer relationship management

Abstract

Disclosed herein are systems, methods, and computer-readable media for troubleshooting based on a probability distribution model. The method for troubleshooting based on a probability distribution model includes establishing a speech-based channel of interaction, establishing at least one non-speech-based channel of interaction, maintaining a probability distribution over time for each of a plurality of component variables describing the state of the product or service and state of the conversation, and troubleshooting a product or service by responding based on the probability distribution.

Claims

1. A method for troubleshooting based on a probability distribution model, the method comprising:

establishing a speech-based conversation with a user;

establishing at least one non-speech-based channel of interaction for obtaining behavioral information associated with a product during the speech-based conversation;

for each of a plurality of component variables, maintaining via a processor a probability distribution for a state of the product based on a state of the speech-based conversation over time and a state of the behavioral information over time; and

troubleshooting the product by responding based on the probability distribution.

2. The method of claim 1, wherein the at least one non-speech-based channel of interaction includes automated responses from the product to troubleshoot.

3. The method of claim 1, wherein the at least one non-speech-based channel of interaction includes observing behavior of the product to troubleshoot.

4. The method of claim 1, wherein the at least one of speech-based conversation and the at least one non-speech-based channel of interaction are established over at least one of a telephone network, a local area network, Internet, and other communications network.

5. The method of claim 1, wherein Bayesian Networks are used to maintain the probability distribution.

6. The method of claim 1, wherein partially observable Markov decision processes (POMDPs) are used to select troubleshooting responses.

7. The method of claim 1, wherein troubleshooting is performed on at least one of a DSL modem, software applications, television, video game console, digital video recorder, computer or components of a computer, cellular phone or other communications device, household appliances, and web page.

8. A system for troubleshooting based on a probability distribution model, the system comprising:

a processor;

a first module configured to control the processor to establish a speech-based conversation with the user;

a second module configured to control the processor to establish at least one non-speech-based channel of interaction for obtaining behavioral information associated with the product during the speech-based conversation;

a third module configured to control the processor to maintain a probability distribution over time for each of a plurality of component variables describing a state of the product based on a state of a conversation over time and a state of the behavioral information over time; and

a fourth module configured to control the processor to troubleshoot the product by responding based on the probability distribution.

9. The system of claim 8, wherein the at least one non-speech-based channel of interaction includes automated responses from the product to troubleshoot.

10. The system of claim 8, wherein the at least one non-speech-based channel of interaction includes observing behavior of the product to troubleshoot.

11. The system of claim 8, wherein at least one of the speech-based conversation and the at least one non-speech-based channel of interaction are established over one of a telephone network, a local area network, Internet, and other communications network.

12. The system of claim 8, wherein Bayesian Networks are used to maintain the probability distribution.

13. The system of claim 8, wherein partially observable Markov decision processes (POMDPs) are used to select troubleshooting responses.

14. The system of claim 8, wherein troubleshooting is performed on at least one of a DSL modem, software applications, television, video game console, digital video recorder, computer or components of a computer, cellular phone or other communications device, household appliances, and web page.

15. A non-transitory computer-readable medium storing a computer program having instructions for troubleshooting based on a probability distribution model, the instructions comprising:

establishing a speech-based conversation with a user;

establishing at least one non-speech-based channel of interaction for obtaining behavioral information associated with the product or service during the sub-based conversation;

for each of a plurality of component variables describing the current state of the product, maintaining a probability distribution over time based on a state of the sub-based conversation over time and a state of the behavioral information over time; and

troubleshooting product by responding based on the probability distribution.

16. The non-transitory computer-readable medium of claim 15, wherein the at least one non-speech-based channel of interaction includes automated responses from the product to troubleshoot.

17. The non-transitory computer-readable medium of claim 15, wherein the at least one non-speech-based channel of interaction includes observing behavior of the product to troubleshoot.

18. The non-transitory computer-readable medium of claim 15, wherein at least one of the speech-based conversation and the at least one non-speech-based channel of interaction are established over one of a telephone network, a local area network, Internet, and other communications network.

19. The non-transitory computer-readable medium of claim 15, wherein Bayesian Networks are used to maintain the probability distribution.

20. The non-transitory computer-readable medium of claim 15, wherein partially observable Markov decision processes (POMDPs) are used to select troubleshooting actions.

21. The non-transitory computer-readable medium of claim 15, wherein troubleshooting is performed on at least one of a DSL modem, software applications, television, video game console, digital video recorder, computer or components of a computer, cellular phone or other communications device, household appliances, and web page.