Systems and methods for automated discrepancy determination, explanation, and resolution

1. A computer-implemented method for autonomously determining and resolving a customer's perceived discrepancy during a customer service interaction, the method comprising:

receiving, by a Natural Language Processing (NLP) device in communication with a discrepancy determination device processor, an incoming communication from a customer;

extracting, by the NLP device, a perceived state and an expected state related to a product or service based on the incoming communication;

acquiring supporting evidence or contrary evidence for the discrepancy;

autonomously determining, by the discrepancy determination device processor, a discrepancy between the perceived and expected state;

verifying, by a rule-based platform of the discrepancy determination device processor, the discrepancy;

extracting, from the incoming communication, one or more verifiable assertions related to the product or service;

verifying, by a rule-based platform of the discrepancy determination device, the one or more verifiable assertions;

generating, by the rule-based platform of the discrepancy determination device processor, response-modifying commands that program the NLP device to generate a response based on the discrepancy;

generating, by the NLP device, and based on the response-modifying commands, the response comprising one or more of:

the one or more verifiable assertions;

a fact pattern response related to the discrepancy; and

a confirmation or correction response related to the discrepancy;

autonomously outputting, for presentation to the customer, the response without involving a human customer service agent; and

ordering a resolution on behalf of the customer based at least in part on the supporting evidence.

2. The method of claim 1 , wherein the response comprises the confirmation or correction response, the method further comprising:

generating, by a machine learning module of the discrepancy determination device, a resolution response, the resolution response comprising an offer to resolve the discrepancy.

3. The method of claim 1 , wherein the response comprises the confirmation or correction response, the method further comprising:

generating, by a machine learning module, a response comprising a request for additional clarification based on a non-verifiable discrepancy.

4. The method of claim 1 , wherein generating the response further comprises generating an explanation for the discrepancy.

5. The method of claim 1 , further comprising:

refining one or more of the rule-based platform and a machine learning module based on intermediary process monitoring of the response.

6. The method of claim 1 , further comprising:

flagging the incoming communication for a call to an external component upon failure of a system comprising the NLP device to generate one or more of the fact pattern response, confirmation, or correction.

7. The method of claim 1 , wherein generating the response further comprises:

identifying the discrepancy in a list of common perceived discrepancies;

retrieving a corresponding explanation from the list; and

adapting the response with the corresponding explanation.

8. The method of claim 1 , further comprising:

generating, based on information about the customer, specific information-eliciting communications for additional dialogue with the customer.

9. The method of claim 1 , further comprising one or more of:

identifying relevant missing information related to the product or service;

abandoning the response and accessing an external component; or

refining the response.

10. A system for determining and resolving a customer's perceived discrepancy during a customer service interaction, the system comprising:

one or more processors;

a discrepancy determination device;

a rule-based platform;

a Natural Language Processing (NLP) device;

a machine learning module; and

memory in communication with the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to:

receive an incoming communication from a customer;

extract, by the NLP device, a perceived state and an expected state of a product or service based on the incoming communication;

acquire supporting evidence or contrary evidence for the discrepancy;

autonomously determine, based on the perceived state and the expected state, and by the discrepancy determination device, a discrepancy between the perceived and expected state of the product or service;

verify, by a rule-based platform, the discrepancy;

extract, from the incoming communication, one or more verifiable assertions related to the product or service;

verify, by the rule-based platform, the one or more verifiable assertions;

generate, by the machine learning module, a response based on the discrepancy, the response comprising one or more of:

the one or more verifiable assertions;

a fact pattern response related to the perceived discrepancy; and

a confirmation or correction of the discrepancy;

refine one or more of the rule-based platform and the machine learning module based on intermediary process monitoring of the response;

autonomously output, for presentation to the customer, the response; and

order a resolution on behalf of the customer based at least in part on the supporting evidence.

11. The system of claim 10 , wherein the instructions further cause the system to generate, by the machine learning module, a resolution response based on the discrepancy, the resolution response comprising an offer to eliminate the discrepancy.

12. The system of claim 10 , wherein the instructions further cause the system to generate, by the machine learning module, a response comprising a request for additional clarification based on a non-verifiable discrepancy.

13. The system of claim 10 , wherein the response comprises an explanation for the discrepancy.

14. The system of claim 10 , wherein the fact pattern response comprises an explanation that there was no actual detected discrepancy.

15. The system of claim 10 , wherein the instructions further cause the system to flag the incoming communication for a call to an external component upon failure of the system to generate one or more of a fact pattern response, a confirmation, or a correction.

16. The system of claim 10 , wherein the response is generated responsive to:

comparing the discrepancy to an entry in a list of common perceived discrepancies;

retrieving a corresponding explanation from the list; and

adapting the response to relate to the incoming communication.

17. The system of claim 10 , wherein the instructions further cause the system to extract and make use of, from the incoming communication, one or more of: an assertion, a desired state, an observed state, and an argument for why the perceived state is incorrect and the expected state is correct.

18. The system of claim 10 , wherein the instructions further cause the system to generate, based on information about the customer, specific information-eliciting communications for additional dialogue with the customer.

19. The system of claim 10 , wherein the instructions further cause the system to perform one or more of:

identify relevant missing information related to the product or service;

abandon the response and access an external component; or

refine the response.