Methods of AI based CRM
In a crowd sourcing approach, responses to customer service inquiries are provided by routing a subset of the inquiries to an independent group of experts. The customer service inquiries are optionally routed to specific experts based on matches between identified subject matter of the inquiries and expertise of the experts. Embodiments include methods of classifying customer service inquiries, training a machine learning system, and/or processing customer service requests.
1. A method of classifying customer service inquiries, the method comprising:
receiving, with an expert management system, a plurality of customer service inquiries;
searching, with cluster detection logic, the customer service inquiries to identify those that fall into previously identified clusters;
using the identified customer service inquiries to train matching logic, with training logic, the matching logic being configured to match the customer service inquires to specific experts;
receiving, with the expert management system, additional customer service inquiries;
matching the additional customer service inquiries to human experts, using the trained matching logic; and
routing, with routing logic, the additional customer service inquiries to the matched human experts.
2. The method of claim 1 , further comprising identifying clusters of customer service inquiries not previously identified.
3. The method of claim 2 , further comprising providing the identified clusters to a human expert for confirmation of the identified clusters.
4. The method of claim 1 , further comprising preparing one or more responses to at least one of the previously identified clusters, the one or more responses being applicable to customer service inquiries within the identified clusters.
5. The method of claim 4 , wherein the prepared one or more responses include response templates.
6. The method of claim 1 , further comprising placing the additional customer service inquiries into the previously identified clusters, wherein the matching of the additional customer service inquiries to the human experts is based on which of the previously identified clusters each of the customer service inquires is placed.
7. The method of claim 1 , wherein the training of the matching logic is based on scored responses to the customer service inquiries.
8. The method of claim 1 , further comprising providing one or more responses to the additional customer service inquiries, the one or more responses being selected based on the previously identified clusters.
9. A method of training a machine learning system, the method comprising:
receiving, with an expert management system, a plurality of customer service inquiries, each of the customer service inquiries being associated with a respective response to that customer service inquiry;
analyzing, with training logic, content of each of the customer service inquires to place the content in a form for automated understanding by the machine learning system;
training, with the training logic, the machine learning system using the customer service inquiries, the associated responses and response scores associated with each of the responses;
testing, with the training logic, the trained machine learning system using additional customer service inquiries, responses to the additional customer service inquires and response scores associated with each of these response scores; and
using the trained machine learning system to process further customer service inquiries, the use being based on the trained machine learning system passing the testing.
10. The method of claim 9 , further comprising, filtering the customer according to whether they are each associated with a positive or negative response.
11. The method of claim 9 , wherein the response scores are based on customer feedback and feedback from human experts.
12. The method of claim 9 , wherein the machine learning system is configured for matching customer service inquiries to human experts.
13. The method of claim 9 , wherein the machine learning system is configured for providing automated responses to customer service inquiries.
14. A method of processing customer service requests, the method comprising:
receiving, with a contact center, a customer service inquiry;
placing, with an expert management system, the customer service inquiry in a queue for processing;
pre-classifying, with the expert management system, the customer service inquiry;
routing, with the expert management system, the customer service inquiry according to the pre-classification of the customer service inquiry;
classifying, with the expert management system, the customer service inquiry according to whether it can be answered by an automated response logic, by an internal expert, or by an external expert, in the alternative;
routing, with routing logic, the customer service inquiry to be answered by the automated response logic, by the internal expert or the external expert, according to the classification of the customer service inquiry; and
generating, with the expert management system, a response to the customer service inquiry using the automated response logic, by the internal expert or the external expert; and
providing, with the expert management system, the response to a source of the customer service inquiry.
15. The method of claim 14 , wherein the pre-classification includes classifying the received customer service inquiries as at least “test sample” and “answerable”.
16. The method of claim 14 , wherein the pre-classification includes classifying the received customer service inquiries as at least “expect negative score” and “answerable”.
17. The method of claim 14 , wherein the pre-classification is performed using a machine learning system trained using customer service inquiries and scored responses to these customer service inquiries.
18. The method of claim 17 , wherein the scored responses are scored by members of the internal or external experts.
19. The method of claim 14 , wherein the classifying of the customer service inquiry includes classifying the customer service inquiry as being a member of a cluster of customer service inquiries requiring similar inquiry responses or having similar topics.
20. The method of claim 14 , wherein the automated response logic is configured to provide a response to the customer service inquiry based on a classification of the customer service inquiry in to a particular cluster.