System and method for query authorization and response generation using machine learning
Systems, methods, and computer-readable storage media for responding to a query using a neural network and natural language processing. If necessary, the system can request disambiguation, then parse the query using a trained machine-learning classifier, resulting in at least one of an identified subject or an identified domain of the text query. The system can determine if the user is authorized to retrieve answers to the query and, if so, retrieve factual data associated with the query. The system can then retrieve a response template, and fill in the template with the retrieved facts. The system can then determine, by executing a machine comprehension model on the filled response template, a probable readability token, portion of text, of at least a portion of the filled response template and, upon identifying that the probable readability is above a threshold, reply to the text query with the at least a portion of the filled response template.
1 . A method comprising:
establishing, by one or more processors, an interface for a chatbot;
receiving, by the one or more processors via the interface for the chatbot, a query from a computing device;
parsing, via the one or more processors executing a trained machine-learning classifier, the query to determine a subject of the query;
selecting, by the one or more processors, based on the subject of the query, a response template from a plurality of response templates stored in a data repository, wherein the response template is selected to provide a response to the query;
filling, via the one or more processors accessing a restricted database, the response template by matching an identifier included in the response template with a key to the restricted database to retrieve at least a portion of restricted data from the restricted database associated with the subject, resulting in a filled response template, wherein the identifier is associated with the at least the portion of the restricted data;
removing, via the one or more processors, one or more portions of the filled response template that are not associated with the subject, resulting in a partially filled response template;
determining, via the one or more processors, based on execution of a comprehension model on the partially filled response template, that a probability associated with the partially filled response template satisfies a threshold; and
providing, via the one or more processors, responsive to the probability associated with the partially filled response template satisfying the threshold, the partially filled response template via the interface of the chatbot.
2 . The method of claim 1 , wherein the trained machine-learning classifier iteratively modifies code executed by the one or more processors upon receiving a threshold number of queries.
3 . The method of claim 2 , wherein the trained machine-learning classifier uses logistic regression between iterations to identify which aspects of the code to modify.
4 . The method of claim 1 , wherein the query is:
received as a speech query; and
converted, by the one or more processors executing a speech-to-text conversion, the speech query into the query.
5 . The method of claim 1 , wherein an identity associated with the computing device identifies at least one of a job title and a clearance level.
6 . The method of claim 1 , wherein the restricted data comprises salary information of other individuals.
7 . A system comprising:
one or more processors; and
a computer-readable storage medium having instructions stored which, when executed by the one or more processors, cause the one or more processors to:
establish an interface for a chatbot;
receive, via the interface for the chatbot, a query from a computing device;
parse, based on execution of a trained machine-learning classifier, the query to determine a subject of the query;
select, based on the subject of the query, a response template from a plurality of response templates stored in a data repository, wherein the response template is selected to provide a response to the query;
fill, based on access to a restricted database, the response template by matching an identifier included in the response template with a key to the restricted database to retrieve at least a portion of restricted data from the restricted database associated with the subject, resulting in a filled response template, wherein the identifier is associated with the at least the portion of the restricted data;
remove one or more portions of the filled response template that are not associated with the subject, resulting in a partially filled response template;
determine, based on execution of a comprehension model on the partially filled response template, that a probability associated with the partially filled response template satisfies a threshold; and
provide, responsive to the probability associated with the partially filled response template satisfying the threshold, the partially filled response template via the interface of the chatbot.
8 . The system of claim 7 , wherein the trained machine-learning classifier iteratively modifies code executed by the one or more processors upon receiving a threshold number of queries.
9 . The system of claim 8 , wherein the trained machine-learning classifier uses logistic regression between iterations to identify which aspects of the code to modify.
10 . The system of claim 7 , wherein the query is:
received as a speech query; and
converted, by the one or more processors based on execution of a speech-to-text conversion, the speech query into the query.
11 . The system of claim 7 , wherein an identity associated with the computing device identifies at least one of a job title and a clearance level.
12 . The system of claim 7 , wherein the restricted data comprises salary information of other individuals.
13 . A non-transitory computer-readable storage medium having instructions stored which, when executed by at least one processor, cause the at least one processor to perform operations comprising:
establishing an interface for a chatbot;
receiving, via the interface for the chatbot, a query from a computing device;
parsing, by executing a trained machine-learning classifier, the query to determine a subject of the query;
selecting, based on the subject of the query, a response template from a plurality of response templates stored in a data repository, wherein the response template is selected to provide a response to the query;
filling, by accessing a restricted database, the response template by matching an identifier included in the response template with a key to the restricted database to retrieve at least a portion of restricted data from the restricted database associated with the subject, resulting in a filled response template, wherein the identifier is associated with the at least the portion of the restricted data;
removing one or more portions of the filled response template that are not associated with the subject, resulting in a partially filled response template;
determining, based on execution of a comprehension model on the partially filled response template, that a probability associated with the partially filled response template satisfies a threshold; and
providing, responsive to the probability associated with the partially filled response template satisfying the threshold, the partially filled response template via the interface of the chatbot.
14 . The non-transitory computer-readable storage medium of claim 13 , wherein the trained machine-learning classifier iteratively modifies code executed by the at least one processor upon receiving a threshold number of queries.
15 . The non-transitory computer-readable storage medium of claim 14 , wherein the trained machine-learning classifier uses logistic regression between iterations to identify which aspects of the code to modify.
16 . The non-transitory computer-readable storage medium of claim 13 , wherein the query is:
received as a speech query; and
converted, by executing a speech-to-text conversion, the speech query into the query.
17 . The non-transitory computer-readable storage medium of claim 13 , wherein an identity associated with the computing device identifies at least one of a job title and a clearance level.