Systems and methods for instructions based messaging inbox using a language model
Various embodiments of this disclosure relate generally to utilizing a machine-learning model to process natural language email instructions. The method comprises receiving, by one or more processors, at least one instruction from a user via an email interface displayed on a user device, the email interface corresponding to an email application, processing, by a trained machine-learning model, the at least one instruction to determine at least one response, wherein processing the at least one instruction includes applying at least one persona of the user to the at least one instruction, performing, by the one or more processors, at least one action or at least one query corresponding to the at least one response, and displaying, by the one or more processors, a confirmation on the email interface of the user device that the at least one response was completed.
1 . A computer-implemented method for utilizing a machine-learning model to process natural language email instructions, the computer-implemented method comprising:
receiving, by one or more processors, at least one instruction from a user via an email interface displayed on a user device, the email interface corresponding to an email application;
processing, by a trained machine-learning model, the at least one instruction to determine at least one response, wherein processing the at least one instruction includes applying at least one persona of the user to the at least one instruction, wherein the trained machine-learning model was previously trained using one or more email datasets and one or more inbox characteristics corresponding to the user to determine the at least one persona corresponding to the user;
performing, by the one or more processors, at least one action or at least one query corresponding to the at least one response; and
displaying, by the one or more processors, a confirmation on the email interface of the user device that the at least one response was completed.
2 . The computer-implemented method of claim 1 , wherein the trained machine-learning model was previously trained using one or more email datasets and one or more inbox characteristics corresponding to the user to determine the at least one persona corresponding to the user.
3 . The computer-implemented method of claim 2 , wherein the one or more inbox characteristics includes at least one of: at least one marked email, at least one email flag, at least one email amount, at least one inbox setting, or at least one inbox rule.
4 . The computer-implemented method of claim 1 , wherein the at least one persona includes at least one of: an inbox organizer persona, a minimalist persona, a priority focused persona, a selective reader persona, a delayed responder persona, a batch processor persona, a social engager persona, an inbox ignorer persona, an information hoarder persona, or an unsubscriber persona.
5 . The computer-implemented method of claim 1 , wherein the at least one instruction includes at least one of: a viewing action, an organizing action, a cleaning action, or a custom action.
6 . The computer-implemented method of claim 1 , the computer-implemented method further comprising:
displaying, by the one or more processors, a user feedback widget on the email interface;
receiving, by the one or more processors, user feedback via the user feedback widget, wherein the user feedback includes a positive feedback indicator, a negative feedback indicator, or a comment; and
retraining, by the one or more processors, the trained machine-learning model based on the user feedback.
7 . The computer-implemented method of claim 1 , wherein the at least one instruction includes at least one natural language query.
8 . The computer-implemented method of claim 1 , the computer-implemented method further comprising:
analyzing, by the one or more processors, the at least one instruction to determine if the at least one instruction meets or surpasses a severity threshold, wherein the severity threshold includes one or more rules corresponding to an impact of the at least one instruction;
in response to determining that the at least one instruction meets or surpasses the severity threshold, capturing, by the one or more processors, a snapshot of the email application; and
storing, by the one or more processors, the snapshot and a corresponding timestamp in one or more databases.
9 . The computer-implemented method of claim 1 , the computer-implemented method further comprising:
receiving, by the one or more processors, a reversion instruction to revert the email application to a previous email application snapshot; and
in response to receiving the reversion instruction, reverting, by the one or more processors, the email application to the previous email application snapshot.
10 . A computer system for utilizing a machine-learning model to process natural language email instructions, the computer system comprising:
a memory having processor-readable instructions stored therein; and
one or more processors configured to access the memory and execute the processor-readable instructions, which when executed by the one or more processors configures the one or more processors to perform a plurality of functions, including functions for:
receiving at least one instruction from a user via an email interface displayed on a user device, the email interface corresponding to an email application;
processing, by a trained machine-learning model, the at least one instruction to determine at least one response, wherein processing the at least one instruction includes applying at least one persona of the user to the at least one instruction, wherein the trained machine-learning model was previously trained using one or more email datasets and one or more inbox characteristics corresponding to the user to determine the at least one persona corresponding to the user;
performing at least one action or at least one query corresponding to the at least one response; and
displaying a confirmation on the email interface of the user device that the at least one response was completed.
11 . The computer system of claim 10 , wherein the trained machine- learning model was previously trained using one or more email datasets and one or more inbox characteristics corresponding to the user to determine the at least one persona corresponding to the user.
12 . The computer system of claim 11 , wherein the one or more inbox characteristics includes at least one of: at least one marked email, at least one email flag, at least one email amount, at least one folder, or at least one inbox rule.
13 . The computer system of claim 10 , the functions further comprising:
displaying a user feedback widget on the email interface;
receiving user feedback via the user feedback widget, wherein the user feedback includes a positive feedback indicator, a negative feedback indicator, or a comment; and
retraining the trained machine-learning model based on the user feedback.
14 . The computer system of claim 10 , the functions further comprising:
analyzing the at least one instruction to determine if the at least one instruction meets or surpasses a severity threshold, wherein the severity threshold includes one or more rules corresponding to an impact of the at least one instruction;
in response to determining that the at least one instruction meets or surpasses the severity threshold, capturing a snapshot of the email application; and
storing the snapshot and a corresponding timestamp in one or more databases.
15 . The computer system of claim 10 , the functions further comprising:
receiving a reversion instruction to revert the email application to a previous email application snapshot; and
in response to receiving the reversion instruction, reverting the email application to the previous email application snapshot.
16 . A non-transitory computer-readable medium containing instructions for utilizing a machine-learning model to process natural language email instructions, the instructions comprising:
receiving at least one instruction from a user via an email interface displayed on a user device, the email interface corresponding to an email application;
processing, by a trained machine-learning model, the at least one instruction to determine at least one response, wherein processing the at least one instruction includes applying at least one persona of the user to the at least one instruction, wherein the trained machine-learning model was previously trained using one or more email datasets and one or more inbox characteristics corresponding to the user to determine the at least one persona corresponding to the user;
performing at least one action or at least one query corresponding to the at least one response; and
displaying a confirmation on the email interface of the user device that the at least one response was completed.
17 . The non-transitory computer-readable medium of claim 16 , the instructions further comprising:
displaying a user feedback widget on the email interface;
receiving user feedback via the user feedback widget, wherein the user feedback includes a positive feedback indicator, a negative feedback indicator, or a comment; and
retraining the trained machine-learning model based on the user feedback.
18 . The non-transitory computer-readable medium of claim 16 , the instructions further comprising:
analyzing the at least one instruction to determine if the at least one instruction meets or surpasses a severity threshold, wherein the severity threshold includes one or more rules corresponding to an impact of the at least one instruction;
in response to determining that the at least one instruction meets or surpasses the severity threshold, capturing a snapshot of the email application; and
storing the snapshot and a corresponding timestamp in one or more databases.
19 . The non-transitory computer-readable medium of claim 16 , the instructions further comprising:
receiving a reversion instruction to revert the email application to a previous email application snapshot; and
in response to receiving the reversion instruction, reverting the email application to the previous email application snapshot.
20 . The non-transitory computer-readable medium of claim 16 , wherein the at least one action includes at least one of: a viewing action, an organizing action, a cleaning action, or a custom action.