Method and system for predicting problematic situations in automated dialog

• US 7,957,970 B1
• Filed: 11/21/2006
• Issued: 06/07/2011
• Est. Priority Date: 11/15/2000
• Status: Expired due to Fees

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Gorin, Allen Louis, Geary, Irene Langkilde, Litman, Diane Judith, Walker, Marilyn Ann, Wright, Jeremy H.
• Primary Examiner(s): Armstrong, Angela A
• Application Number:
  US11/562,080
• Time in Patent Office: 1,659 Days
• Field of Search: 704/251, 704/255, 704/257, 704/270, 704/270.1, 704/275
• US Class Current: 704/257
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: H04M 3/4936 Speech interaction details speech recognition per se G10L15/00

Abstract

The invention concerns a system and method of predicting problematic dialogs in an automated dialog system based on the user'"'"'s input communications. The method may include determining whether a probability of conducting a successful dialog with the user exceeds a first threshold. The successful dialog may be defined as a dialog exchange between an automated dialog system and the user that results in at least one of processing of the user'"'"'s input communication and routing the user'"'"'s input communication. The method may further operate such that if the first threshold is exceeded, further dialog is conducted with the user. Otherwise, the user may be directed to a human for assistance. In another possible embodiment, the method operates as above except that if the probability exceeds a second threshold, the second threshold being higher than the first, then further dialog is conducted with the user using the current dialog strategy. However, if the probability falls between a first threshold and a second threshold, the dialog strategy may be adapted in order to improve the chances of conducting a successful dialog with the user. This process may be cumulative. In particular, the first dialog exchange may be stored in a database. Then, a second dialog exchange is conducted with the user. As a result, a second determination is made as whether successful dialog can be conducted based on the stored first exchange and the current second exchanges. This cumulative process may continue using a third and fourth exchange, if necessary.

Claims

1. A method comprising:

conducting a dialog exchange between an automated dialog system and a user, the dialog exchange including at least two dialog system prompts and two user input opportunities; and

determining whether a probability threshold is exceeded that relates to a prediction that continuing the dialog exchange with the user will converge and succeed to a stage consistent with the user'"'"'s intent or diverge and fail, wherein determining whether the probability threshold is exceeded is based, at least in part, training data on multiple dialog turns in a corpus of dialogs tagged for success or failure after at least two dialog prompts and two user inputs.

2. The method of claim 1, wherein the dialog training data includes at least one of dialog classification models and extracted dialog features.

3. The method of claim 2, wherein the extracted dialog features are derived from recognition, understanding and dialog data.

4. The method of claim 1, wherein if the probability threshold is exceeded, further dialog is conducted with the user.

5. The method of claim 1, wherein the probability threshold is a first threshold, the method further comprising:

determining whether a probability of conducting a successful dialog with the user exceeds a second threshold, the second threshold being greater than the probability threshold, wherein if the second threshold is exceeded, further dialog is conducted with the user using a current dialog strategy.

6. The method of claim 5, wherein if the second threshold is not exceeded, further dialog is conducted with the user using an adapted dialog strategy.

7. The method of claim 6, wherein the adapted dialog strategy includes one of prompting the user with choices and prompting the user to confirm recognition and understanding data.

8. The method of claim 1, wherein the user'"'"'s input communication includes at least one of verbal and nonverbal communications.

9. The method of claim 8, wherein the nonverbal communications include at least one of gestures, body movements, head movements, non-responses, text, keyboard entries, keypad entries, mouse clicks, DTMF codes, pointers, stylus, cable set-top box entries, graphical user interface entries, and touchscreen entries.

10. The method of claim 1, wherein the method is used for customer care purposes.

11. The method of claim 1, wherein the probability is determined using recognition and understanding data derived from the user'"'"'s input communication.

12. The method of claim 1, wherein if the probability threshold is not exceeded, the user is routed to a human for assistance.

13. The method of claim 1, wherein the dialog training data is stored in a dialog training database.

14. A system comprising:

a processor;

a module configured to control the processor to conduct a dialog exchange between an automated dialog system and a user, the dialog exchange including at least two dialog system prompts and two user input opportunities; and

a module configured to determine whether a probability threshold is exceeded that relates to a prediction that continuing the dialog exchange with the user will converge and succeed to a stage consistent with the user'"'"'s intent or diverge and fail, wherein determining whether the probability threshold is exceeded is based, at least in part, training data on multiple dialog turns in a corpus of dialogs tagged for success or failure after at least two dialog prompts and two user inputs.

15. The system of claim 14, wherein the dialog training data includes at least one of dialog classification models and extracted dialog features.

16. The system of claim 14, wherein if the probability threshold is not exceeded, the user is routed to a human for assistance.

17. The system of claim 14, wherein if the probability threshold is succeeded, further dialog is conducted with the user.

18. The system of claim 14, wherein the probability threshold is a first threshold, the method further comprising:

determining whether a probability of conducting a successful dialog with the user exceeds a second threshold, the second threshold being greater than the probability threshold, wherein if the second threshold is exceeded, further dialog is conducted with the user using a current dialog strategy.

19. The system of claim 18, wherein if the second threshold is not exceeded, further dialog is conducted with the user using an adapted dialog strategy.

20. The system of claim 19, wherein the adapted dialog strategy includes one of prompting the user with choices and prompting the user to confirm recognition and understanding data.

21. The system of claim 14, wherein the user'"'"'s input communication includes at least one of verbal and nonverbal communications.

22. The system of claim 21, wherein the nonverbal communications include at least one of gestures, body movements, head movements, non-responses, text, keyboard entries, keypad entries, mouse clicks, DTMF codes, pointers, stylus, cable set-top box entries, graphical user interface entries, and touchscreen entries.