System and method for IVR signature detection and event management

• US 8,160,211 B1
• Filed: 07/12/2010
• Issued: 04/17/2012
• Est. Priority Date: 03/10/2004
• Status: Active Grant

• Current Assignee: Concentrix CVG Customer Management Group, Inc.
• Original Assignee: Convergys Customer Management Group Incorporated (Convergys Corporation)
• Inventors: Bittner, Stephen Nicholas
• Primary Examiner(s): SING, SIMON P
• Application Number:
  US12/834,120
• Time in Patent Office: 645 Days
• Field of Search: 379/72, 379/88.18, 379/201.12
• US Class Current: 379/72
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements
• CPC Class Codes: H04M 3/527 Centralised call answering arrangements not requiring operator intervention

Abstract

An IVR signature detection and event management system facilitates real-time discovery and notification of abnormal behavior within an IVR system. The system collects metrics associated with usage of the IVR system. The system uses the collected metrics to define a baseline. Having defined a baseline, the system collects evaluation metrics for comparison to the baseline. Where the evaluation metrics deviate from the baseline beyond a threshold deviation, the system detects aberrance. The system triggers an event in response to the detected aberrance. The event may be a message, one or more corrective actions, or the like.

Claims

1. A method of providing dialog system monitoring, the method comprising:

(a) capturing a first set of metrics associated with use of a dialog system by a plurality of users, the metrics of the first set comprising at least one of;

(i) dialog navigation sequence,

(ii) dialog termination point, or

(iii) number turns per dialog;

(b) defining a baseline sample based on the captured first set of metrics;

(c) capturing a second set of metrics associated with use of the dialog system by a plurality of users, the metrics of the second set comprising at least one of;

(i) dialog navigation sequence,

(ii) dialog termination point, or

(iii) number turns per dialog;

(d) defining a comparison sample based on the second set of metrics;

(e) comparing the comparison sample to the baseline sample; and

(f) triggering an event when the step of comparing the comparison sample to the baseline sample reveals aberrance.

2. The method of claim 1, wherein the step of comparing the comparison sample to the baseline sample comprises determining a deviation of the comparison sample from the baseline sample.

3. The method of claim 2, wherein the step of comparing the comparison sample to the baseline sample further comprises determining whether the deviation exceeds a threshold deviation.

4. The method of claim 1, wherein the step of triggering an event comprises sending a message including notice of the aberrance.

5. The method of claim 4, wherein the notice includes a priority indicator.

6. The method of claim 1, wherein the step of triggering an event comprises performing an automated diagnostic function.

7. The method of claim 1, wherein the step of triggering an event comprises sending a message to at least one of a systems analyst or systems maintenance personnel.

8. The method of claim 1, wherein the step of triggering an event comprises sending a message to a caller, the message comprising at least one of additional options or instructions.

9. The method of claim 1, wherein the baseline sample comprises a statistical distribution of a first set of dialog navigation sequences.

10. The method of claim 9, wherein the comparison sample comprises a statistical distribution of a second set of dialog navigation sequences.

11. The method of claim 1, further comprising updating the baseline sample.

12. The method of claim 1, wherein the step of capturing a second set of metrics is performed subsequent to the step of capturing a first set of metrics.

13. The method of claim 1, further comprising defining a plurality of baseline samples, wherein each baseline sample of the plurality of baseline samples corresponds to a type of comparison sample.

14. The method of claim 13, wherein each baseline sample of the plurality of baseline samples relates to at least one metric that is not related to the other baselines of the plurality of baseline samples.

15. The method of claim 1, wherein the dialog system is configured to receive input comprising spoken statements by users.

16. The method of claim 15 wherein the dialog system comprises an IVR system.

17. A method of providing real-time dialog system monitoring comprising:

(a) capturing metrics associated with use of a dialog system by a plurality of users, said metrics comprising a first set of dialog navigation sequences;

(b) defining a baseline sample based on the first set of dialog navigation sequences;

(c) capturing subsequent metrics associated with use of the dialog system by a plurality of users subsequent to the step of defining a baseline sample, wherein the subsequent metrics comprise a second set of dialog navigation sequences;

(d) defining a comparison sample based on the second set of dialog navigation sequences;

(e) comparing the comparison sample to the baseline sample; and

(f) triggering an event in real time when the step of comparing the comparison sample to the baseline sample reveals aberrance.

18. The method of claim 17, wherein the baseline sample comprises a statistical distribution based on the first set of dialog navigation sequences.

19. The method of claim 18, wherein the comparison sample comprises a statistical distribution based on the second set of dialog navigation sequences.

20. A method of providing dialog system monitoring comprising:

(a) capturing a plurality of metrics associated with use of a dialog system by a plurality of users, the metrics including at least one of;

(i) dialog navigation sequence,

(ii) dialog termination point, or

(iii) number turns per dialog;

(b) defining a baseline based at least in part on the captured plurality of metrics

(c) defining a sample based at least in part on the captured plurality of metrics

(d) comparing the sample to the baseline; and

(e) triggering an event when the act of comparing reveals aberrance.