Dynamic recommendation of routing rules for contact center use
View Patent ↗A machine learning system and method for contact center use. Activities associated with a plurality of contact centers are monitored and a knowledge base is updated based on the monitored activities. An outcome for a particular contact center may be predicted based on monitored interactions for the particular contact center, and based on information in the knowledge base. An output is then generated based on the predicted outcome.
1. A machine learning method for contact center use, the machine learning method comprising:
monitoring, by a processor, activities associated with a plurality of contact centers;
extracting a pattern based on the monitored activities;
identifying customer satisfaction predictors from the extracted pattern;
updating, by the processor, a knowledge base based on the monitored activities;
monitoring, by the processor, interaction for a particular contact center;
predicting, by the processor, an outcome for the particular contact center based on the interaction monitored for the particular contact center, and the customer satisfaction predictors identified from the extracted pattern;
generating, by the processor, a recommended routing strategy for a routing point based on the predicted outcome, wherein the recommended routing strategy includes computer instructions that indicate how interactions arriving at the routing point are to be routed to resources of the contact center; and
establishing, by an edge device, a communication session between a particular resource of the contact center and a particular end user device based on execution of the routing strategy.
2. The method of claim 1 , wherein the monitored activities are customer interactions handled by the plurality of contact centers.
3. The method of claim 1 , wherein the monitored activities include an amount of time a call is put on hold prior to being routed to an agent.
4. The method of claim 1 , wherein the monitored activities include a number of transfers made during a call.
5. The method of claim 1 , wherein the predicted outcome is feedback from a customer associated with the monitored interaction indicative of lack of satisfaction by the customer.
6. The method of claim 1 , wherein the predicted outcome is abandonment of the call.
7. The method of claim 1 further comprising:
validating the outcome for the particular contact center via simulation based on prior interaction data.
8. The method of claim 1 , wherein the knowledge base includes a machine learning model, wherein the updating includes adjusting parameters to the machine learning model based on the monitored activities.
9. The method of claim 8 , wherein the outcome for the particular contact center is predicted based on the machine learning model.
10. A machine learning system for contact center use, the machine learning system comprising:
one or more processors; and
one or more memory devices coupled to the one or more processors and having program instructions stored therein, the one or more processors being configured to execute the program instructions, the program instructions comprising:
monitoring activities associated with a plurality of contact centers;
extracting a pattern based on the monitored activities;
identifying customer satisfaction predictors from the extracted pattern;
updating a knowledge base based on the monitored activities;
monitoring interaction for a particular contact center;
predicting an outcome for the particular contact center based on the interaction monitored for the particular contact center, and the customer satisfaction predictors identified from the extracted pattern; and
generating a recommended routing strategy for a routing point based on the predicted outcome, wherein the recommended routing strategy includes computer instructions that indicate how interactions arriving at the routing point are to be routed to resources of the contact center for the particular contact center and other ones of the plurality of contact centers; and
an edge device coupled to the one or more processors configured to establish a communication session between a particular resource of the contact center and a particular end user device based on execution of the routing strategy.
11. The system of claim 10 , wherein the monitored activities are customer interactions handled by the plurality of contact centers.
12. The system of claim 10 , wherein the monitored activities include an amount of time a call is put on hold prior to being routed to an agent.
13. The system of claim 10 , wherein the monitored activities include a number of transfers made during a call.
14. The system of claim 10 , wherein the predicted outcome is feedback from a customer associated with the monitored interaction indicative of lack of satisfaction by the customer.