Method and system for capturing and storing machine learned quantitative classification of natural language data

1 . A method for facilitating qualitative assessment of natural language data via artificial intelligence, the method being implemented by at least one processor, the method comprising:

receiving, by the at least one processor via an application programming interface, a first input from at least one source, the first input including the natural language data;

parsing, by the at least one processor via at least one model, the first input to identify at least one linguistic element of a natural language, wherein the at least one model includes a machine learning and pattern recognition model, wherein the machine learning and pattern recognition model includes an artificial neural network;

identifying, by the at least one processor via the at least one model, subject matter of the first input;

delineating, by the at least one processor via the at least one model, the first input into respective contextually distinct input groups based on the identified subject matter;

determining, by the at least one processor using the at least one model, at least one confidence score for each respective contextually distinct input group, the at least one confidence score relating to a clarity level of the natural language data that is determined by using a word sense disambiguation model that identifies a context of the at least one linguistic element of the first input, wherein the at least one model is trained using an assessment model that is based on an error rate associated with the determined at least one confidence score for each respective contextually distinct input group;

for each respective contextually distinct input group, determining, by the at least one processor using the at least one model, whether the at least one confidence score exceeds a predetermined threshold;

determining, by the at least one processor and based on the first input, a respective corresponding downstream application for a respective corresponding contextually distinct input group when the at least one confidence score exceeds a predetermined threshold, wherein the respective corresponding downstream application relates to a classification and metadata tagging operation for processing the first input;

processing, by the at least one processor via the respective corresponding downstream application, each respective contextually distinct input group, wherein a result of the processing of each respective contextually distinct input group is used to further train the at least one model;

generating, by the at least one processor using the at least one model, at least one request for additional information when the at least one confidence score is below the predetermined threshold, the at least one request including a prompt in a natural language format; and

transmitting, by the at least one processor via the application programming interface, the at least one request back to the at least one source;

receiving, by the at least one processor via the application programming interface, a second input from the at least one source in response to the at least one request, the second input including the additional information;

determining, by the at least one processor using the at least one model, at least one new confidence score based on the first input and the second input; and

determining, by the at least one processor using the at least one model, whether the at least one new confidence score exceeds the predetermined threshold,

wherein the first input is automatically separated into a plurality of constituent inputs of the first input based on contextual information provided by the second input and the at least one new confidence score, each of the plurality of constituent inputs including contextually distinct subject matter.

2 . The method of claim 1 , further comprising:

associating, by the at least one processor, the first input with the at least one confidence score; and

transmitting, by the at least one processor, the first input and the associated at least one confidence score to the respective corresponding downstream application.

3 . The method of claim 2 , wherein the respective corresponding downstream application includes network location information for the transmitting of the first input and the associated at least one confidence score.

4 . The method of claim 1 , wherein the clarity level includes at least one from among a clear level, a high clarity level, a medium clarity level, and a low clarity level, the clarity level corresponding to a predicted coherency characteristic of the natural language data.

5 . The method of claim 1 , wherein the at least one model further includes at least one from among a natural language processing model, a machine learning model, a mathematical model, a process model, and a data model.

6 . A computing device configured to implement an execution of a method for facilitating qualitative assessment of natural language data via artificial intelligence, the computing device comprising:

a processor;

a memory; and

a communication interface coupled to each of the processor and the memory,

wherein the processor is configured to:

receive, via an application programming interface, a first input from at least one source, the first input including the natural language data;

parse, via at least one model, the first input to identify at least one linguistic element of a natural language, wherein the at least one model includes a machine learning and pattern recognition model, wherein the machine learning and pattern recognition model includes an artificial neural network;

identify, via the at least one model, subject matter of the first input;

delineate, via the at least one model, the first input into respective contextually distinct input groups based on the identified subject matter;

determine, by using the at least one model, at least one confidence score for each respective contextually distinct input group, the at least one confidence score relating to a clarity level of the natural language data that is determined by using a word sense disambiguation model that identifies a context of the at least one linguistic element of the first input, wherein the at least one model is trained using an assessment model that is based on an error rate associated with the determined at least one confidence score for each respective contextually distinct input group;

for each respective contextually distinct input group, determine, by using the at least one model, whether the at least one confidence score exceeds a predetermined threshold;

determine, based on the first input, a respective corresponding downstream application for a respective corresponding contextually distinct input group when the at least one confidence score exceeds a predetermined threshold, wherein the respective corresponding downstream application relates to a classification and metadata tagging operation for processing the first input;

process, via the respective corresponding downstream application, each respective contextually distinct input group, wherein a result of the processing of each respective contextually distinct input group is used to further train the at least one model;

generate, by using the at least one model, at least one request for additional information when the at least one confidence score is below the predetermined threshold, the at least one request including a prompt in a natural language format;

transmit, via the application programming interface, the at least one request back to the at least one source;

receive, via the application programming interface, a second input from the at least one source in response to the at least one request, the second input including the additional information;

determine, via the at least one model, at least one new confidence score based on the first input and the second input; and

determine, via the at least one model, whether the at least one new confidence score exceeds the predetermined threshold,

wherein the first input is automatically separated into a plurality of constituent inputs of the first input based on contextual information provided by the second input and the at least one new confidence score, each of the plurality of constituent inputs including contextually distinct subject matter.

7 . The computing device of claim 6 , wherein the processor is further configured to:

associate the first input with the at least one confidence score; and

transmit the first input and the associated at least one confidence score to the respective corresponding downstream application.

8 . The computing device of claim 7 , wherein respective corresponding downstream application includes network location information for the transmitting of the first input and the associated at least one confidence score.

9 . The computing device of claim 6 , wherein the clarity level includes at least one from among a clear level, a high clarity level, a medium clarity level, and a low clarity level, the clarity level corresponding to a predicted coherency characteristic of the natural language data.

10 . The computing device of claim 6 , wherein the at least one model further includes at least one from among a natural language processing model, a machine learning model, a mathematical model, a process model, and a data model.

11 . A non-transitory computer readable storage medium storing instructions for facilitating qualitative assessment of natural language data via artificial intelligence, the storage medium comprising executable code which, when executed by a processor, causes the processor to:

receive, via an application programming interface, a first input from at least one source, the first input including the natural language data;

parse, via at least one model, the first input to identify at least one linguistic element of a natural language, wherein the at least one model includes a machine learning and pattern recognition model, wherein the machine learning and pattern recognition model includes an artificial neural network;

identify, via the at least one model, subject matter of the first input;

delineate, via the at least one model, the first input into respective contextually distinct input groups based on the identified subject matter;

determine, by using the at least one model, at least one confidence score for each respective contextually distinct input group, the at least one confidence score relating to a clarity level of the natural language data that is determined by using a word sense disambiguation model that identifies a context of the at least one linguistic element of the first input, wherein the at least one model is trained using an assessment model that is based on an error rate associated with the determined at least one confidence score for each respective contextually distinct input group;

for each respective contextually distinct input group, determine, by using the at least one model, whether the at least one confidence score exceeds a predetermined threshold;

determine, based on the first input, a respective corresponding downstream application for a respective corresponding contextually distinct input group when the at least one confidence score exceeds a predetermined threshold, wherein the respective corresponding downstream application relates to a classification and metadata tagging operation for processing the first input;

process, via the respective corresponding downstream application, each respective contextually distinct input group, wherein a result of the processing of each respective contextually distinct input group is used to further train the at least one model;

generate, by using the at least one model, at least one request for additional information when the at least one confidence score is below the predetermined threshold, the at least one request including a prompt in a natural language format;

transmit, via the application programming interface, the at least one request back to the at least one source;

receive, via the application programming interface, a second input from the at least one source in response to the at least one request, the second input including the additional information;

determine, via the at least one model, at least one new confidence score based on the first input and the second input; and

determine, via the at least one model, whether the at least one new confidence score exceeds the predetermined threshold,

wherein the first input is automatically separated into a plurality of constituent inputs of the first input based on contextual information provided by the second input and the at least one new confidence score, each of the plurality of constituent inputs including contextually distinct subject matter.