Matching a customer and customer representative dynamically based on a customer representative's past performance

1. A method for customer representative matching, the method comprising:

separating a received chat transcript into a set of triplets, wherein a machine learning model utilizes a trained intent and entity classifier to tag each triplet in the set of triplets with one or more pre-defined tags;

generating one or more multi-dimensional success vectors based on the one or more pre-defined tags;

aggregating the one or more multi-dimensional success vectors;

receiving at least one business priority;

training a second machine learning model, which utilizes TF-IDF clustering, on a set of historical chat transcript data and metadata, and the aggregated one or more multi-dimensional success vectors;

querying the trained second machine learning model to match the customer to at least one customer representative by filtering out at least one past utterance which contains similar metadata to a current utterance and by identifying a most successful utterance which is semantically similar to a new customer question; and

revealing a match by outputting a response to the customer and connecting, via a chat interface, the customer and the at least one customer representative.

2. The method of claim 1 , wherein aggregating the one or more multi-dimensional success vectors further comprises:

assigning a weight to diminish a value of an older answer based on the older answer's distance from a present time and difference from a present success metric.

3. The method of claim 1 , wherein receiving the at least one business priority further comprises:

assigning a weight to each of the at least one business priorities, wherein a high business priority is given a high weight and a low business priority is given a low weight; and

increasing the weight of the low business priority based on a user input.

4. The method of claim 1 , wherein training the machine learning model to match the customer to the at least one customer representative further comprises:

using the one or more multi-dimensional success vectors to learn the at least one customer representative's success in each dimension; and

leveraging a learned success when matching the customer to the at least one customer representative.

5. The method of claim 1 , wherein querying the trained machine learning model to match the customer to the at least one customer representative further comprises:

querying the trained machine learning model with a customer question, wherein the customer question contains one or more dimensions of the multi-dimensional success vector;

determining the at least one customer representative with the most success in the one or more dimensions; and

providing the customer with a highly rated response of the at least one customer representative.

6. The method of claim 1 , wherein querying the trained machine learning model to match the customer to the at least one customer representative further comprises:

querying the trained machine learning model with a customer question, wherein the customer question contains one or more dimensions of the multi-dimensional success vector;

determining the at least one customer representative with the most success in the one or more dimensions; and

connecting the customer and the at least one customer representative.

7. A computer system for customer representative matching, comprising:

one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage medium, and program instructions stored on at least one of the one or more tangible storage medium for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising:

separating a received chat transcript into a set of triplets, wherein a machine learning model utilizes a trained intent and entity classifier to tag each triplet in the set of triplets with one or more pre-defined tags;

generating one or more multi-dimensional success vectors based on the one or more pre-defined tags;

aggregating the one or more multi-dimensional success vectors;

receiving at least one business priority;

training a second machine learning model, which utilizes TF-IDF clustering, on a set of historical chat transcript data and metadata, and the aggregated one or more multi-dimensional success vectors;

querying the trained second machine learning model to match the customer to at least one customer representative by filtering out at least one past utterance which contains similar metadata to a current utterance and by identifying a most successful utterance which is semantically similar to a new customer question; and

revealing a match by outputting a response to the customer and connecting, via a chat interface, the customer and the at least one customer representative.

8. The computer system of claim 7 , wherein aggregating the one or more multi-dimensional success vectors further comprises:

assigning a weight to diminish a value of an older answer based on the older answer's distance from a present time and difference from a present success metric.

9. The computer system of claim 7 , wherein receiving the at least one business priority further comprises:

assigning a weight to each of the at least one business priorities, wherein a high business priority is given a high weight and a low business priority is given a low weight; and

increasing the weight of the low business priority based on a user input.

10. The computer system of claim 7 , wherein training the machine learning model to match the customer to the at least one customer representative further comprises:

using the one or more multi-dimensional success vectors to learn the at least one customer representative's success in each dimension; and

leveraging a learned success when matching the customer to the at least one customer representative.

11. The computer system of claim 7 , wherein querying the trained machine learning model to match the customer to the at least one customer representative further comprises:

querying the trained machine learning model with a customer question, wherein the customer question contains one or more dimensions of the multi-dimensional success vector;

determining the at least one customer representative with the most success in the one or more dimensions; and

providing the customer with a highly rated response of the at least one customer representative.

12. The computer system of claim 7 , wherein querying the trained machine learning model to match the customer to the at least one customer representative further comprises:

querying the trained machine learning model with a customer question, wherein the customer question contains one or more dimensions of the multi-dimensional success vector;

determining the at least one customer representative with the most success in the one or more dimensions; and

connecting the customer and the at least one customer representative.

13. A computer program product for customer representative matching, comprising:

one or more non-transitory computer-readable storage media and program instructions stored on at least one of the one or more tangible storage media, the program instructions executable by a processor to cause the processor to perform a method comprising:

separating a received chat transcript into a set of triplets, wherein a machine learning model utilizes a trained intent and entity classifier to tag each triplet in the set of triplets with one or more pre-defined tags;

generating one or more multi-dimensional success vectors based on the one or more pre-defined tags;

aggregating the one or more multi-dimensional success vectors;

receiving at least one business priority;

training a second machine learning model, which utilizes TF-IDF clustering, on a set of historical chat transcript data and metadata, and the aggregated one or more multi-dimensional success vectors;

querying the trained second machine learning model to match the customer to at least one customer representative by filtering out at least one past utterance which contains similar metadata to a current utterance and by identifying a most successful utterance which is semantically similar to a new customer question; and

revealing a match by outputting a response to the customer and connecting, via a chat interface, the customer and the at least one customer representative.

14. The computer program product of claim 13 , wherein aggregating the one or more multi-dimensional success vectors further comprises:

assigning a weight to diminish a value of an older answer based on the older answer's distance from a present time and difference from a present success metric.

15. The computer program product of claim 13 , wherein receiving the at least one business priority further comprises:

assigning a weight to each of the at least one business priorities, wherein a high business priority is given a high weight and a low business priority is given a low weight; and

increasing the weight of the low business priority based on a user input.

16. The computer program product of claim 13 , wherein training the machine learning model to match the customer to the at least one customer representative further comprises:

using the one or more multi-dimensional success vectors to learn the at least one customer representative's success in each dimension; and

leveraging a learned success when matching the customer to the at least one customer representative.

17. The computer program product of claim 13 , wherein querying the trained machine learning model to match the customer to the at least one customer representative further comprises:

querying the trained machine learning model with a customer question, wherein the customer question contains one or more dimensions of the multi-dimensional success vector;

determining the at least one customer representative with the most success in the one or more dimensions; and

providing the customer with a highly rated response of the at least one customer representative.

18. The computer program product of claim 13 , wherein querying the trained machine learning model to match the customer to the at least one customer representative further comprises:

querying the trained machine learning model with a customer question, wherein the customer question contains one or more dimensions of the multi-dimensional success vector;

determining the at least one customer representative with the most success in the one or more dimensions; and

connecting the customer and the at least one customer representative.