Message routing in a contact center
The present disclosure provides, among other things, a method of managing contacts in a contact center, the method including: receiving a text-based communication from a customer of the contact center; analyzing the text-based communication to determine a relevancy associated with the text-based communication; based on the analysis, determining a relevancy level to assign to the text-based communication; tagging the text-based communication with a relevancy tag that identifies the determined relevancy level; updating a priority associated with assigning the text-based communication to an agent of the contact center based on the relevancy tag; assigning the text-based communication to the agent of the contact center; enabling a machine learning process to analyze a database of text-based communications; and updating a data model used to automatically tag text-based communications with relevancy tags based on the analysis performed by the machine learning process.
1. A method of managing contacts in a contact center, comprising:
receiving a communication in the contact center after the communication has passed through a network border device and had a rules-based SPAM filter applied thereto, wherein the network border device initially identifies the communication as not SPAM;
analyzing the communication that has passed through the network border device with an additional filter that passes the communication through a machine learning process;
receiving an output from the machine learning process that identifies the communication as SPAM even though the communication passed through the network border device and was identified as not SPAM;
tagging the communication with an indicator that identifies the communication as SPAM; and
in response to tagging the communication with the indicator, automatically deleting the communication and/or automatically placing the communication in a sandbox thereby preventing the communication from being routed to an agent.
2. The method of claim 1 , further comprising:
receiving an agent input that confirms the communication is appropriately tagged as SPAM; and
updating, based on receiving the agent input, a data model used by the machine learning process to make future decisions on communication filtering.
3. The method of claim 1 , wherein the communication comprises a text-based communication.
4. The method of claim 3 , wherein the text-based communication comprises an email.
5. The method of claim 1 , wherein the communication is automatically deleted in response to tagging the communication with the indicator.
6. The method of claim 1 , wherein the communication is automatically placed in the sandbox in response to tagging the communication with the indicator.
7. The method of claim 1 , wherein the machine learning process utilizes a data model that receives, as an input, the communication and that provides, in response to processing the input, the output that identifies the communication as SPAM.
8. The method of claim 7 , wherein the data model comprises at least one of a Decision Tree, a Support Vector Machine (SVM), a Nearest Neighbor, and a Bayesian classifier.
9. The method of claim 1 , wherein the machine learning process is trained on a first type of text-based communication and used to make filtering decisions on a second type of text-based communication.
10. A system, comprising:
a processor; and
computer memory storing data thereon that enables the processor to:
receive a communication after the communication has passed through a network border device and had a rules-based SPAM filter applied thereto, wherein the network border device initially identifies the communication as not SPAM;
analyze the communication that has passed through the network border device with an additional filter that passes the communication through a machine learning process;
receive an output from the machine learning process that identifies the communication as SPAM even though the communication passed through the network border device and was identified as not SPAM;
tag the communication with an indicator that identifies the communication as SPAM; and
in response to tagging the communication with the indicator, automatically delete the communication and/or automatically place the communication in a sandbox thereby preventing the communication from being routed to an agent.
11. The system of claim 10 , wherein the machine learning process utilizes a data model that receives, as an input, the communication and that provides, in response to processing the input, the output that identifies the communication as SPAM.
12. The system of claim 10 , wherein the communication comprises an email and wherein the email is automatically deleted in response to tagging the communication with the indicator.
13. A method, comprising:
receiving a communication after the communication has passed through a network border device and had a rules-based filter applied thereto, wherein the network border device initially identifies the communication as wanted;
analyzing the communication that has passed through the network border device with an additional filter that passes the communication through a machine learning process;
receiving an output from the machine learning process that identifies the communication unwanted even though the communication passed through the network border device and was identified as wanted;
tagging the communication with an indicator that identifies the communication as unwanted; and
in response to tagging the communication with the indicator, automatically processing the communication to prevent further routing of the communication to a user.
14. The method of claim 13 , wherein automatically processing the communication comprises failing to route the communication across a communication network.
15. The method of claim 13 , wherein automatically processing the communication comprises placing the communication in a sandbox.
16. The method of claim 13 , wherein automatically processing the communication comprises automatically deleting the communication.
17. The method of claim 13 , wherein the communication comprises a text-based communication.
18. The method of claim 13 , wherein the communication comprises a voice communication.
19. The method of claim 13 , wherein the communication comprises a video communication.