Customer relationship management system based on electronic conversations

Patent metadata

Field	Value
Publication number	US-9538007-B1
Application number	US-201514930965-A
Country	US
Kind code	B1
Filing date	Nov 3, 2015
Priority date	Nov 3, 2015
Publication date	Jan 3, 2017
Grant date	Jan 3, 2017

Abstract

The disclosed embodiments illustrate methods and systems for managing a conversation between a first user and a customer care agent. The method includes extracting one or more features from a first conversation between said first user and said customer care agent. The first conversation corresponds to an ongoing conversation over an electronic communication medium. The method includes determining a score for a feature. The score indicates a deviation of a value of said feature from an average of values of said feature determined from historical conversations involving said first user. The method includes aggregating said score for each of said one or more features of said first conversation. Thereafter, the method includes redirecting said first conversation of said first user to a third user during said first conversation based on said aggregation. The third user manages said first conversation between said first user and said customer care agent.

First claim

What is claimed is:

1. A method for managing, by a computing device, a conversation between a first user and a second user, said method comprising:
extracting, by one or more microprocessors in said computing device, one or more features from a first conversation between said first user and said second user, wherein said first conversation corresponds to an ongoing conversation over an electronic communication medium;
for a feature from said one or more extracted features:
determining, by said one or more microprocessors, a score for said feature, wherein said score indicates a deviation of a value of said feature from an average of values of said feature determined from one or more historical conversations, extracted from a database server over a communication network, involving said first user;
aggregating, by said one or more microprocessors, said score for each of said one or more features of said first conversation; and
redirecting, by said one or more microprocessors, said first conversation of said first user to a third user during said first conversation based on said aggregated score, resulting in reduced time spent on identification of one or more queries of said first user.

2. The method of claim 1 further comprising extracting, by said one or more microprocessors, said one or more features from said one or more historical conversations involving said first user, by utilizing one or more speech or text processing techniques.

3. The method of claim 2 further comprising extracting, by said one or more microprocessors, values of said feature of each of said one or more historical conversations during a duration of each of said one or more historical conversations.

4. The method of claim 3 further comprising determining, by said one or more microprocessors, said average of values of said feature of each of said one or more historical conversations by utilizing an average computation technique, wherein said average computation technique comprises at least a threshold based sequential time delay embedding, or a group means over time technique.

5. The method of claim 1 , wherein said first conversation is not redirected to said third user when said aggregated score is less than a threshold value.

6. The method of claim 1 , wherein said first conversation is further redirected based on an availability of said third user and said determined aggregated score is greater than a threshold value.

7. The method of claim 1 , wherein said one or more features comprise at least one of said first user's turnaround time, said second user's turnaround time, a turn length of said first user, a turn length of said second user, sentiments of said first user and said second user, a speaking rate of said second user, or said second user's response delay.

8. The method of claim 1 , wherein said score for said feature is determined by utilizing at least an anomaly detection technique, or a deviation computation technique.

9. The method of claim 1 , wherein said first user corresponds to a customer, a user, or a client.

10. The method of claim 1 , wherein said second user corresponds to an automated agent, or a customer care representative.

11. The method of claim 1 , wherein said third user corresponds to an expert, a supervisor, or a manager.

12. The method of claim 1 , wherein said electronic communication medium comprises at least one of a short message service (SMS), a chat, social networks, an email, or a phone.

13. A system for managing, by a computing device a conversation between a first user and a second user, said system comprising:
one or more microprocessors in said computing device operable to:
extract one or more features from a first conversation between said first user and said second user, wherein said first conversation corresponds to an ongoing conversation over an electronic communication medium;
for a feature from said one or more extracted features:
determine a score for said feature, wherein said score indicates a deviation of a value of said feature from an average of values of said feature determined from one or more historical conversations, extracted from a database server over a communication network, involving said first user;
aggregate said score for each of said one or more features of said first conversation; and
redirect said first conversation of said first user to a third user during said first conversation based on said aggregated score, resulting in reduced time spent on identification of one or more queries of said first user.

14. The system of claim 13 , wherein said one or more microprocessors are further operable to extract said one or more features from said one or more historical conversations involving said first user by utilizing one or more speech or text processing techniques.

15. The system of claim 14 , wherein said one or more microprocessors are further operable to extract values of said feature of each of said one or more historical conversations during a duration of each of said one or more historical conversations.

16. The system of claim 15 , wherein said one or more microprocessors are further operable to determine said average of values of said feature of each of said one or more historical conversations by utilizing an average computation technique, wherein said average computation technique comprises at least a threshold based sequential time delay embedding, or a group means over time technique.

17. The system of claim 13 , wherein said first conversation is not redirected to said third user when said aggregated score is less than a threshold value.

18. The system of claim 13 , wherein said first conversation is redirected based at least on an availability of said third and said determined aggregated score is greater than a threshold value.

19. The system of claim 13 , wherein said one or more features comprise at least one of said first user's turnaround time, said second user's turnaround time, a turn length of said first user, a turn length of said second user, sentiments of said first user and said second user, a speaking rate of said second user, or said second user's response delay.

20. The system of claim 13 , wherein said score for said feature is determined by utilizing at least an anomaly detection technique, or a deviation computation technique.

21. The system of claim 13 , wherein said first user corresponds to a customer, a user, or a client.

22. The system of claim 13 , wherein said second user corresponds to an automated agent, or a customer care representative.

23. The system of claim 13 , wherein said third user corresponds to an expert, a supervisor, or a manager.

24. The system of claim 13 , wherein said electronic communication medium comprises at least one of a short message service (SMS), a chat, social networks, an email, or a phone.

25. A computer program product for use with a computer, the computer program product comprising a non-transitory computer readable medium, wherein the non-transitory computer readable medium stores a computer program code for managing a conversation between a first user and a second user, wherein the computer program code is executable by one or more microprocessors to:
extract, by one or more microprocessors in a computing device, one or more features from a first conversation between said first user and said second user, wherein said first conversation corresponds to an ongoing conversation over an electronic communi

Assignees

• Xerox Corp

Inventors

• Mariappan Ragunathan
• Roy Shourya

Classifications

• G06Q30/015

  Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk · CPC title

• H04M3/5232

  Call distribution algorithms · CPC title

• H04M3/51

  Centralised call answering arrangements requiring operator intervention {, e.g. call or contact centers for telemarketing} · CPC title

• H04M3/5183

  Call or contact centers with computer-telephony arrangements · CPC title

• H04M3/5175 Primary

  Call or contact centers supervision arrangements · CPC title