System and method for predicting customer satisfaction

• US 9,176,950 B2
• Filed: 12/12/2012
• Issued: 11/03/2015
• Est. Priority Date: 12/12/2012
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Whitten, Michael L., Perkner, John M., Sistrunk, Jacob Keikiokelii
• Primary Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US13/712,600
• Publication Number:
  US 20140163961A1
• Time in Patent Office: 1,056 Days
• Field of Search: 704/9
• US Class Current: 1/1
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

A system includes a memory and a processor communicatively coupled to the memory. The memory stores interaction data associated with an interaction between a customer and an associate of an entity. The processor is operable to determine, from the interaction data, one or more keywords in the interaction between the customer and the associate, determine an order of the one or more keywords, and determine a grouping of the one or more keywords. The processor determines, based on the determined keywords, order, and grouping, a perception of the entity by the customer, the determination of the perception of the entity occurring in real-time after the interaction between the customer and the associate.

Claims

1. A system, comprising:

a telephone operable to facilitate an interaction between a customer and an associate of an entity;

one or more memory units; and

a processor communicatively coupled to the telephone and the one or more memory units and operable to;

record the interaction between the customer and the associate of the entity and store the interaction in the one or more memory units;

using the recorded interaction, automatically generate interaction data and store the interaction data in the one or more memory units, wherein the interaction data comprises;

speaker separation data indicating whether the customer or the associate of the entity is speaking in the interaction; and

talk-over data indicating one or more instances when the customer and the associate of the entity are speaking over each other;

retrieve the stored interaction data and determine, from the retrieved interaction data, one or more keywords in the interaction between the customer and the associate;

determine, from the retrieved interaction data, an order of the one or more keywords;

determine, from the retrieved interaction data, a grouping of the one or more keywords;

determine, based on the keywords, order, grouping, speaker separation data, and talk-over data, a perception of the entity by the customer, the determination of the perception of the entity occurring in real-time after the interaction between the customer and the associate.

2. The system of claim 1, wherein the interaction data further comprises one or more of:

a word/phrase index;

speech-to-text data;

emotion data;

one or more images; and

silence data.

3. The system of claim 1, the processor further operable to determine, from the interaction data, one or more emotion indications in the interaction between the customer and the associate, wherein the determination of the perception of the entity by the customer is further based on the one or more emotion indications.

4. The system of claim 1, the processor further operable to predict, based on the determination of the perception of the entity by the customer, one or more future actions by the customer.

5. The system of claim 1, the processor further operable to provide, based on the determination of the perception of the entity by the customer, feedback to the associate.

6. The system of claim 1, wherein the perception of the entity by the customer is determined using Boolean logic.

7. A method comprising:

automatically generating, by a computer system, interaction data associated with a recorded interaction between a customer and an associate of an entity, wherein the interaction data comprises;

speaker separation data indicating whether the customer or the associate of the entity is speaking in the interaction; and

talk-over data indicating one or more instances when the customer and the associate of the entity are speaking over each other;

determining, by the computer system from the interaction data, one or more keywords in the interaction between the customer and the associate;

determining, by the computer system from the interaction data, an order of the one or more keywords;

determining, by the computer system from the interaction data, a grouping of the one or more keywords; and

determining, by the computer system based on the keywords, order, grouping, speaker separation data, and talk-over data, a perception of the entity by the customer, the determination of the perception of the entity occurring in real-time after the interaction between the customer and the associate.

8. The method of claim 7, wherein the interaction data comprises data associated with a telephone conversation between the customer and the associate.

9. The method of claim 7, wherein the interaction data comprises data associated with an Internet chat session between the customer and the associate.

10. The method of claim 7, wherein the interaction data further comprises one or more of:

a word/phrase index;

speech-to-text data;

emotion data;

one or more images; and

silence data.

11. The method of claim 7, further comprising determining, from the interaction data, one or more emotion indications in the interaction between the customer and the associate, wherein the determination of the perception of the entity by the customer is further based on the one or more emotion indications.

12. The method of claim 7, further comprising predicting, based on the determination of the perception of the entity by the customer, one or more future actions by the customer.

13. The method of claim 7, further comprising providing, based on the determination of the perception of the entity by the customer, feedback to the associate.

14. One or more computer-readable non-transitory storage media embodying software this is operable when executed by one or more computer systems to:

automatically generate interaction data associated with a recorded interaction between a customer and an associate of an entity, wherein the interaction data comprises;

speaker separation data indicating whether the customer or the associate of the entity is speaking in the interaction; and

talk-over data indicating one or more instances when the customer and the associate of the entity are speaking over each other;

determine, from the interaction data, one or more keywords in the interaction between the customer and the associate;

determine, from the interaction data, an order of the one or more keywords;

determine, from the interaction data, a grouping of the one or more keywords; and

determine, based on the keywords, order, grouping, speaker separation data, and talk-over data, a perception of the entity by the customer, the determination of the perception of the entity occurring in real-time after the interaction between the customer and the associate.

15. The media of claim 14, wherein the interaction data comprises data associated with a telephone conversation between the customer and the associate or an Internet chat session between the customer and the associate.

16. The media of claim 14, wherein the interaction data further comprises one or more of:

a word/phrase index;

speech-to-text data;

emotion data;

one or more images; and

silence data.

17. The media of claim 14, the software further operable to determine, from the interaction data, one or more emotion indications in the interaction between the customer and the associate, wherein the determination of the perception of the entity by the customer is further based on the one or more emotion indications.

18. The media of claim 14, the software further operable to predict, based on the determination of the perception of the entity by the customer, one or more future actions by the customer.