Self-learning adaptive routing system

• US 10,038,788 B1
• Filed: 05/09/2017
• Issued: 07/31/2018
• Est. Priority Date: 05/09/2017
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Khalatian, Igor
• Primary Examiner(s): Nguyen, Khai N
• Application Number:
  US15/590,543
• Time in Patent Office: 448 Days
• Field of Search: 37926509
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: H04L 41/5064 Customer relationship management
• CPC Class Codes: H04L 65/1069 Session establishment or de-establishment
• CPC Class Codes: H04L 65/60 Network streaming of media packets
• CPC Class Codes: H04M 3/5191 interacting with the Internet
• CPC Class Codes: H04M 3/5233 Operator skill based call distribution
• CPC Class Codes: H04M 7/0042 where the data service is a text-based messaging service H04M7/0054 takes precedence

Abstract

A method for self-learning adaptive routing can include receiving a contact along with a sequence of events from a customer via a graphical user interface on a web page within a browser window. A universal resource locator (“URL”) of the web page can be captured and mapped with the sequence of events to the contact. A matching routing rule can be used to route the contact to an appropriate customer service queue. An agent associated with the customer service queue can view a co-browsing preview of the customer'"'"'s desktop, which the agent can use to transfer the contact to a different customer service queue. A machine learning algorithm can create a new routing rule based on the URL of the web page, the sequence of events, the co-browsing preview, the second routing, and the determination that the second agent associated with the second customer service queue resolved the contact.

Claims

1. A method for improving a Customer Relationship Management (“

CRM”

) system, comprising;

receiving, at the CRM system, a contact from a customer, the contact being initiated through a graphical user interface on a web page within a browser window;

receiving, at the CRM system, a sequence of events captured from an interaction of the customer with the web page within the browser window;

capturing, by the CRM system, a Universal Resource Locator (“

URL”

) of the web page;

mapping, by the CRM system, the URL of the web page and the sequence of events to the contact;

querying, by the CRM system, a rule store having a plurality of routing rules;

matching, by the CRM system, a first routing rule of the plurality of routing rules to the contact;

routing, by the CRM system, the contact to a first customer service queue based on the first routing rule;

providing, by the CRM system, a co-browsing preview of the browser window to a computer of a first agent associated with the first customer service queue;

detecting, by the CRM system, a second routing of the contact from the first customer service queue to a second customer service queue;

determining, by the CRM system, that a second agent associated with the second customer service queue resolved the contact; and

creating, by the CRM system, a new routing rule using a machine learning algorithm based on the URL of the web page, the sequence of events, the co-browsing preview, the second routing, and the determination that the second agent associated with the second customer service queue resolved the contact.

2. The method of claim 1, wherein the sequence of events is a first sequence of events, the method further comprising:

receiving, at the CRM system, a second contact from a second customer, the second contact being initiated through the graphical user interface on the web page within a second browser window;

receiving, at the CRM system, a second sequence of events captured from a second interaction of the second customer with the web page within the second browser window, wherein the second sequence of events is equivalent to the first sequence of events;

mapping, by the CRM system, the URL of the web page and the second sequence of events with the second contact; and

routing, by the CRM system, the second contact to the second customer service queue using the new routing rule.

3. The method of claim 1, wherein the machine learning algorithm comprises:

detecting, by the CRM system, a plurality of contacts that have been initiated through the graphical user interface on the web page and have been routed from the first customer service queue to the second customer service queue;

determining, by the CRM system, that a threshold number of the plurality of contacts were resolved by one or more agents associated with the second customer service queue; and

in response to determining that the threshold number of contacts were resolved by the one or more agents associated with the second customer service queue, enabling, by the CRM system, the new routing rule to route contacts initiated through the graphical user interface of the web page to the second customer service queue.

4. The method of claim 3, wherein the new routing rule is a first new routing rule and wherein the machine learning algorithm further comprises:

detecting, by the CRM system, a second plurality of contacts that have been initiated through the graphical user interface on the web page and have been routed from the second customer service queue to a third customer service queue;

in response to determining that the threshold number of contacts from the second plurality of contacts were resolved by one or more agents associated with the third customer service queue, creating, by the CRM system, a second new routing rule to route contacts initiated through the graphical user interface of the web page to the third customer service queue;

enabling, by the CRM system, the second new routing rule to route contacts initiated through the graphical user interface of the web page to the third customer service queue; and

disabling, by the CRM system, the first new routing rule.

5. The method of claim 1, wherein the first agent manually routes the contact to the second customer service queue based on the co-browsing preview of the web page with the customer.

6. The method of claim 1, wherein at least one event of the sequence of events comprises an error code.

7. The method of claim 1, wherein:

a first event of the sequence of events is a login attempt;

a second event of the sequence of events is an error code corresponding to a failed login attempt; and

the second customer service queue is a security team queue.

8. A system, comprising:

a processor; and

a memory having stored thereon instructions, which when executed by the processor, cause the processor to;

receive a contact from a customer, the contact being initiated through a graphical user interface on a web page within a browser window;

receive a sequence of events captured from an interaction of the customer with the web page within the browser window;

capture a Universal Resource Locator (“

URL”

) of the web page;

map the URL of the web page and the sequence of events with the contact;

query a rule store having a plurality of routing rules;

match a first routing rule of the plurality of routing rules to the contact;

route the contact to a first customer service queue based on the first routing rule;

provide a co-browsing preview of the browser window to a computer of a first agent associated with the first customer service queue;

detect a second routing of the contact from the first customer service queue to a second customer service queue;

determine that a second agent associated with the second customer service queue resolved the contact; and

create a new routing rule using a machine learning algorithm based on the URL of the web page, the sequence of events, the co-browsing preview, the second routing, and the determination that the second agent associated with the second customer service queue resolved the contact.

9. The system of claim 8, wherein the sequence of events is a first sequence of events and wherein the memory has stored thereon further instructions, which when executed by the processor, cause the processor to:

receive a second contact from a second customer, the second contact being initiated through the graphical user interface on the web page within a second browser window;

receive a second sequence of events captured from a second interaction of the second customer with the web page within the second browser window, wherein the second sequence of events is equivalent to the first sequence of events;

map the URL of the web page and the second sequence of events with the second contact; and

route the second contact to the second customer service queue using the new routing rule.

10. The system of claim 8, wherein the machine learning algorithm comprises instructions, which when executed by the processor, cause the processor to:

detect a plurality of contacts that have been initiated through the graphical user interface on the web page and have been routed from the first customer service queue to the second customer service queue;

determine that a threshold number of the plurality of contacts were resolved by one or more agents associated with the second customer service queue; and

in response to determining that the threshold number of contacts were resolved by the one or more agents associated with the second customer service queue, enable the new routing rule to route contacts initiated through the graphical user interface of the web page to the second customer service queue.

11. The system of claim 10, wherein the new routing rule is a first new routing rule and wherein the machine learning algorithm comprises further instructions, which when executed by the processor, cause the processor to:

detect a second plurality of contacts that have been initiated through the graphical user interface on the web page and have been routed from the second customer service queue to a third customer service queue;

in response to determining that the threshold number of contacts from the second plurality of contacts were resolved by one or more agents associated with the third customer service queue, create a second new routing rule to route contacts initiated through the graphical user interface of the web page to the third customer service queue;

enable the second new routing rule to route contacts initiated through the graphical user interface of the web page to the third customer service queue; and

disable the first new routing rule.

12. The system of claim 8, wherein the first agent manually routes the contact to the second customer service queue based on the co-browsing preview of the web page with the customer.

13. The system of claim 8, wherein at least one event from the sequence of events comprises an error code.

14. The system of claim 8, wherein:

a first event of the sequence of events is a login attempt;

a second event of the sequence of events is an error code corresponding to a failed login attempt; and

the second customer service queue is a security team queue.

15. A computer-readable memory device having stored thereon a set of instructions, which when executed by a processor, causes the processor to:

receive a contact from a customer, the contact being initiated through a graphical user interface on a web page within a browser window;

receive a sequence of events captured from an interaction of the customer with the web page within the browser window;

capture a Universal Resource Locator (“

URL”

) of the web page;

map the URL of the web page and the sequence of events with the contact;

query a rule store having a plurality of routing rules;

match a first routing rule of the plurality of routing rules to the contact;

route the contact to a first customer service queue based on the first routing rule;

provide a co-browsing preview of the browser window to a computer of a first agent associated with the first customer service queue;

detect a second routing of the contact from the first customer service queue to a second customer service queue;

determine that a second agent associated with the second customer service queue resolved the contact; and

create a new routing rule using a machine learning algorithm based on the URL of the web page, the sequence of events, the co-browsing preview, the second routing, and the determination that the second agent associated with the second customer service queue resolved the contact.

16. The computer-readable memory device of claim 15, wherein the sequence of events is a first sequence of events, and wherein the computer-readable memory device having stored thereon further instructions, which when executed by the processor, cause the processor to:

receive a second contact from a second customer, the second contact being initiated through the graphical user interface on the web page within a second browser window;

receive a second sequence of events captured from a second interaction of the second customer with the web page within the second browser window, wherein the second sequence of events is equivalent to the first sequence of events;

map the URL of the web page and the second sequence of events with the second contact; and

route the second contact to the second customer service queue using the new routing rule.

17. The computer-readable memory device of claim 15, wherein the machine learning algorithm comprises instructions, which when executed by the processor, cause the processor to:

detect a plurality of contacts that have been initiated through the graphical user interface on the web page and have been routed from the first customer service queue to the second customer service queue;

determine that a threshold number of the plurality of contacts were resolved by one or more agents associated with the second customer service queue; and

in response to determining that the threshold number of contacts were resolved by the one or more agents associated with the second customer service queue, enable the new routing rule to route contacts initiated through the graphical user interface of the web page to the second customer service queue.

18. The computer-readable memory device of claim 17, wherein the new routing rule is a first new routing rule and wherein the machine learning algorithm comprises further instructions, which when executed by the processor, cause the processor to:

detect a second plurality of contacts that have been initiated through the graphical user interface on the web page and have been routed from the second customer service queue to a third customer service queue;

in response to determining that the threshold number of contacts from the second plurality of contacts were resolved by one or more agents associated with the third customer service queue, create a second new routing rule to route contacts initiated through the graphical user interface of the web page to the third customer service queue;

enable the second new routing rule to route contacts initiated through the graphical user interface of the web page to the third customer service queue; and

disable the first new routing rule.

19. The computer-readable memory device of claim 15, wherein the first agent manually routes the contact to the second customer service queue based on the co-browsing preview of the web page with the customer.

20. The computer-readably memory device of claim 15, wherein at least one event of the sequence of events comprises an error code.