CPC G06Q 40/08 (2013.01) [G06F 40/103 (2020.01); G06F 40/284 (2020.01); G06F 40/40 (2020.01); G06N 20/00 (2019.01)] | 14 Claims |
1. A platform-agnostic digital assistant apparatus for electronic communication with a user in a natural language format, the apparatus comprising:
a processor; and
a memory, wherein the memory is electronically and communicatively coupled with the processor and storing instructions configuring the processor to:
receive user data from a storage, wherein the storage comprises user informational contents;
establish an electronic communication channel with a computing device of an entity, wherein the entity computing device is in communication with the user for providing services to the user, the entity computing device including an interface for receiving digital communication from the user;
utilize the communication channel to receive a first information related to a user from the entity, wherein the first information comprises communication exchanged between the entity and the user in natural language format for initiating an action on behalf of the user;
extract at least one user datum from the first information and the user informational contents in the storage;
determine a user communication style from the user datum, wherein determining the user communication style further comprises:
analyzing, using the digital assistant, the user datum to determine a first communication style;
generating, using the digital assistant, first communication style training data based on the first communication style, wherein the first communication style training data comprises correlations between exemplary language elements which correspond to the first communication style, wherein the first communication style training data is labeled in accordance with the first communication style, wherein the language elements comprise at least one of words, tokens of words, and components of the language of the first information;
training a communication style machine learning model using only the first communication style training data, wherein the training is dynamic as the first information is received, and wherein the model is configured to generate a dynamic response in the determined first communication style; and
determining, using the communication style machine learning model, an appropriate natural language response in the first communication style, wherein the response is a function of the correlated exemplary language elements;
modify the informational contents of the storage based on at least one of the user datum and the first information; and
transmit a response as a function of the appropriate natural language response and the first communication style to the user.
|