| CPC G06F 40/56 (2020.01) | 18 Claims |
|
1. A system for generating real-time dynamic conversational responses during conversational interactions using machine learning models based on historic intents for a plurality of users and user-specific interactions, the system comprising:
one or more processors; and
a non-transitory, computer-readable medium comprising instructions that, when executed by the one or more processors, cause operations comprising:
receiving, from a first user, a first user action during a conversational interaction with a user interface, wherein the first user action comprises a text-based communication, wherein the conversational interaction comprises an interactive exchange of text messages between the first user and a mobile application;
generating a first feature input based on the first user action, wherein the first feature input is an embedding, based on natural language processing, of contents of the first user action;
inputting the first feature input into a first machine learning model, wherein the first machine learning model comprises a model trained to identify user intents based on embeddings of existing intent parent classes and labeled child intent classes corresponding to one or more of the existing intent parent and child classes, wherein the first machine learning model comprises a hierarchical local binary classification system, and wherein training the first machine learning model comprises using a local classifier that uses preexisting embeddings as feed forward classifiers to predict labels for its corresponding parent class or child class;
receiving a first output from the first machine learning model, wherein the first output comprises a label classification for the first input;
determining whether the first output corresponds to a new intent; and
generating for display, at the user interface during the conversational interaction, a first dynamic conversational response based on determining that the first feature output corresponds to the new intent.
|