| CPC G06F 40/284 (2020.01) [G06F 3/017 (2013.01); G06F 40/40 (2020.01)] | 13 Claims |
|
1. A three-tiered selection system for selecting either a contextual prediction processing or a classical prediction processing for providing a response to an electronic user input, the three-tiered selection system comprising:
a receiver operating at an entity-specific interactive voice and text response system (“IVR”), said entity-specific IVR comprising at least one hardware processor and hardware memory, said receiver operable to receive the electronic user input from a software application operating on a hardware device associated with a user; and
a selection processor, said selection processor comprising:
a first tier, said first tier operable to:
identify that the electronic user input is a first user input within a conversation and/or the electronic user input comprises a gesture entered by the user at the software application;
initiate a classical analysis on the electronic user input, said classical analysis configured to:
receive a plurality of identifying details relating to the user in addition to the electronic user input;
electronically build, based on an action-topic ontology, said action-topic ontology being a language interpretable by the entity-specific IVR, said action-topic ontology including a plurality of entity-specific verbs, a plurality of entity-specific topics, a conversation frame corresponding to the electronic user input and the plurality of identifying details; and
electronically communicate the conversation frame to a software code element within the entity-specific IVR, said software code element operable to generate a response to the electronic user input;
identify, at the software code element, using the classical analysis, a classical response to the electronic user input; and
electronically communicate the classical response as a text output and/or as an audio output to user via the software application;
a second tier, said second tier operable to:
initiate the classical analysis on the electronic user input;
identify, using the classical analysis, the classical response to the electronic user input;
identify a classical confidence value for the classical response to the electronic user input;
identify that the classical confidence value is above a predetermined confidence value; and
electronically communicate the classical response as a text output and/or as an audio output to the user via the software application; and
a third tier, said third tier operable to:
initiate a contextual analysis on the electronic user input when the classical confidence value is below the predetermined confidence value, said contextual analysis operable to transform the electronic user input into a contextual user input based on two or more user inputs included in the conversation, the contextual analysis configured to:
receive one or more previous user inputs and a second plurality of details, said second plurality of details comprising identifying details relating to the user and data relating to the one or more previous user inputs, said data relating to the one or more previous user inputs comprising a position of each of the previous user inputs within the conversation;
receive the conversation frame;
merge the conversation frame with the second plurality of details to generate a target conversation frame, said target conversation frame structured to provide an answer to the user input;
electronically communicate the target conversation frame to the software code element;
identify, at the software code element, a contextual response to the contextual user input;
identify a contextual confidence value for the contextual response to the contextual user input;
compare the contextual confidence value to the classical confidence value;
identify a sentiment analysis score for the contextual response and a sentiment analysis score for the classical response when the contextual confidence value is within a predetermined value window from the classical confidence value;
compare the sentiment analysis score of the classical response to the sentiment analysis score of the contextual response;
electronically communicate the contextual response as a text output and/or as an audio output to the user via the software application when the sentiment analysis score for the classical response is greater than the sentiment analysis score for the contextual response; and
electronically communicate the classical response as a text output and/or as an audio output to the user via the software application when the sentiment analysis score for the contextual response is greater than the sentiment analysis score for the classical response.
|