System and methods for determining trends in electronic communications

• US 10,419,611 B2
• Filed: 09/29/2008
• Issued: 09/17/2019
• Est. Priority Date: 09/28/2007
• Status: Active Grant

• Current Assignee: Mattersight Corporation (Nice Ltd)
• Original Assignee: Mattersight Corporation (Nice Ltd)
• Inventors: Conway, Kelly, Yengoyan, Alan, Brown, Douglas, Gustafson, David, Danson, Christopher
• Primary Examiner(s): Matar, Ahmad F.
• Assistant Examiner(s): Intavong, Jirapon
• Application Number:
  US12/286,169
• Publication Number:
  US 20090103709A1
• Time in Patent Office: 4,005 Days
• Field of Search: 37925606, 7042701
• CPC Class Codes: H04M 3/42221 Conversation recording systems at the subscriber's set H04M1/656
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements

Abstract

The invention relates to a method and system for analyzing an electronic communication, more particularly, to analyzing telephonic communications between customers and a contact center to determine and display the most relevant communications to an organization or business.

Claims

1. A node comprising:

a processor; and

a non-transitory computer readable medium operably coupled thereto, which includes logic adapted to analyze a plurality of electronic communications between one or more customers of an organization and agents of a contact center comprising;

instructions that, when executed, apply linguistic analysis to each electronic communication that occurs over a first predetermined time period to identify segments including words related to business or non-business categories within each electronic communication;

instructions, that when executed, apply non-linguistic analysis to each electronic communication that occurs over the first predetermined time period, wherein the non-linguistic analysis includes at least one of a timing of each segment of each electronic communication, a determination of a position of each segment within each electronic communication, a determination of a quantity of each segment within each electronic communication, and a determination of an interaction between the one or more customers and agents within each electronic segment;

instructions that, when executed, identify a pattern in each electronic communication based on the linguistic analysis and the non-linguistic analysis over the first predetermined time period;

instructions that, when executed, determine a frequency for the pattern in each electronic communication over the first predetermined time period;

instructions that, when executed, identify a first subset of communications from the plurality of electronic communications that are relevant based on the identified pattern and the determined frequency for the pattern over the first predetermined time period;

instructions that, when executed, identify a second subset of communications that exhibit distress from the first subset of communications;

instructions that, when executed, identify a trend in the second subset of communications over the first predetermined time period based on the determined frequency for the pattern and presence of distress;

instructions that, when executed, display the identified trend on a graphical user interface; and

instructions that, when executed, transmit the second subset of communications to relevant personnel of the organization, the second subset of communications being the most relevant communications relative to the plurality of electronic communications.

2. The node of claim 1, wherein the logic further comprises:

instructions that, when executed, identify a pattern in each electronic communication that occurs over a second predetermined time period;

instructions that, when executed, determine a frequency for the pattern in each electronic communication over the second predetermined time period;

instructions that, when executed, identify a subset of communications from the plurality of electronic communications that are relevant based on the identified pattern and the determined frequency over the second predetermined time period; and

instructions that, when executed, identify a trend in the subset of communications over the second predetermined time period.

3. The node of claim 2, wherein the first and second predetermined time periods are sequential and do not overlap.

4. The node of claim 2, wherein the logic further comprises instructions, that when executed, generate a recommended action based on a comparison of a first plurality of identified patterns from the first predetermined time period to a second plurality of identified patterns from the second predetermined time period.

5. The node of claim 4, wherein the logic further comprises instructions, that when executed, transmit the recommended action to a communication distributor.

6. The node of claim 1, wherein the instructions that, when executed, identify the pattern comprise instructions, that when executed, establish a frequency threshold, and include a pattern that exceeds the frequency threshold in the identified pattern.

7. The node of claim 1, which further comprises a routing engine configured to send an automated message to an external source based on the identified trend.

8. The node of claim 1, wherein the logic further comprises at least one of:

instructions that, when executed, determine how relevant each electronic communication is to different areas of a business organization using a relevance filter, and

instructions that, when executed, determine how important each electronic communication is to different areas of a business organization using an importance filter.

9. The node of claim 8, wherein the logic further comprises instructions that, when executed, generate an interactive vignette to accompany identified relevant and identified important electronic communications.

10. The node of claim 1, wherein the logic further comprises:

instructions that, when executed, identify a linguistic behavioral signifier of at least one of the one or more customers and agents;

instructions that, when executed, compare the identified linguistic behavioral signifier to a database including a plurality of behavioral signifiers;

instructions that, when executed, apply a predetermined algorithm to the identified linguistic behavioral signifier; and

instructions that, when executed, determine a personality type associated with the identified linguistic behavioral signifier based on the comparison of the identified linguistic behavioral signifier to the database of behavioral signifiers and the application of the predetermined algorithm to the identified linguistic behavioral signifier.

11. The node of claim 1, wherein the logic further comprises:

instructions that, when executed, categorize each electronic communication as a type of communication.

12. The node of claim 11, wherein the type of communication is one of a service type communication, a retention type communication, and a sales type communication.

13. The node of claim 1, wherein the non-linguistic analysis further comprises a spectral analysis of an audio waveform of each electronic communication.

14. A computer-implemented method for analyzing electronic communications between one or more customers of an organization and a contact center having one or more agents, the method being implemented by a non-transitory computer readable medium adapted to control a computer and comprising a plurality of code segments, the method comprising:

recording a plurality of electronic communications;

applying linguistic analysis to each recorded electronic communication that occurs over a first predetermined time period to identify segments including words related to business or nonbusiness categories within each electronic communication;

applying non-linguistic analysis to each electronic communication that occurs over the first predetermined time period, wherein the non-linguistic analysis includes at least one of timing of each segment of each electronic communication, determining a position of each segment within each electronic communication, determining a quantity of each segment within each electronic communication, and determining an interaction between the one or more customers and agents within each electronic communication;

identifying a pattern in each recorded electronic communication based on the linguistic analysis and the non-linguistic analysis over the first predetermined time period;

determining a frequency for the pattern in each recorded electronic communication over the first predetermined time period;

identifying a first subset of communications from the plurality of recorded electronic communications that are relevant based on the identified pattern and the determined frequency for the pattern over the first predetermined time period;

identifying a second subset of communications that exhibit distress from the first subset of communications;

identifying a trend in the second subset of communications over the first predetermined time period based on the determined frequency for the pattern and presence of distress;

displaying the identified trend on a graphical user interface; and

transmitting the second subset of communications to relevant personnel of the organization, the second subset of communications being the most relevant communications relative to the plurality of electronic communications.

15. The computer-implemented method of claim 14, which further comprises:

identifying a pattern in each recorded electronic communication that occurs over a second predetermined time period;

determining a frequency for the pattern in each recorded electronic communication over the second predetermined time period;

identifying a subset of communications from the plurality of electronic communications that are relevant based on the identified pattern and the determined frequency over the second predetermined time period; and

identifying a trend in the subset of communications over the second predetermined time period.

16. The computer-implemented method of claim 15, wherein the first and second predetermined time periods are sequential and do not overlap.

17. The computer-implemented method of claim 15, which further comprises generating a recommended action based on a comparison of a first plurality of identified patterns from the first predetermined time period to a second plurality of identified patterns from the second predetermined time period.

18. The computer-implemented method of claim 17, which further comprises transmitting the recommended action to a communication distributor.

19. The computer-implemented method of claim 14, wherein identifying the pattern comprises establishing a frequency threshold, and including a pattern that exceeds the frequency threshold in the identified pattern.

20. The computer-implemented method of claim 14, which further comprises sending an automated message to an external source based on the identified trend.

21. A non-transitory machine-readable medium comprising a plurality of machine-readable instructions which, when executed by one or more processors, are adapted to cause the one or more processors to perform a method which comprises:

recording a plurality of electronic communications between one or more customers of an organization and a contact center having one or more agents;

applying linguistic analysis to each recorded electronic communication that occurs over a first predetermined time period to identify segments including words related to business or nonbusiness categories within each electronic communication;

applying non-linguistic analysis to each electronic communication that occurs over the first predetermined time period, wherein the non-linguistic analysis includes at least one of timing of each segment of each electronic communication, determining a position of each segment within each electronic communication, determining a quantity of each segment within each electronic communication, and determining an interaction between the one or more customers and agents within each electronic communication;

identifying a pattern in each recorded electronic communication based on the linguistic analysis and the non-linguistic analysis over the first predetermined time period;

determining a frequency for the pattern in each recorded electronic communication over the first predetermined time period;

identifying a first subset of communications from the plurality of recorded electronic communications that are relevant based on the identified pattern and the determined frequency for the pattern over the first predetermined time period;

identifying a second subset of communications that exhibit distress from the first subset of communications;

identifying a trend in the second subset of communications over the first predetermined time period based on the determined frequency for the pattern;

generating a recommended action based on the identified trend;

displaying the recommended action to a user; and

transmitting the second subset of communications to relevant personnel of the organization, the second subset of communications being the most relevant communications relative to the plurality of electronic communications.

22. The non-transitory machine-readable medium of claim 21, wherein the method further comprises:

identifying a pattern in each recorded electronic communications that occurs over a second predetermined time period;

determining a frequency for the pattern in each recorded electronic communication over the second predetermined time period;

identifying a subset of communications from the plurality of recorded electronic communications that are relevant based on the identified pattern and the determined frequency over the second predetermined time period; and

identifying a trend in the subset of communications over the second predetermined time period.

23. The non-transitory machine-readable medium of claim 22, wherein the first and second predetermined time periods are sequential and do not overlap.

24. The non-transitory machine-readable medium of claim 22, wherein generating the recommended action is based on a comparison of a first plurality of identified patterns from the first predetermined time period to a second plurality of identified patterns from the second predetermined time period.

25. The non-transitory machine-readable medium of claim 21, wherein identifying the pattern further comprises establishing a frequency threshold, and including a pattern that exceeds the frequency threshold in the identified pattern.

26. The non-transitory machine-readable medium of claim 21, wherein the method further comprises sending an automated message to an external source based on the identified trend.