Training an artificial intelligence engine to predict responses for determining appropriate action

Patent metadata

Field	Value
Publication number	US-12154128-B2
Application number	US-202217661558-A
Country	US
Kind code	B2
Filing date	May 1, 2022
Priority date	May 1, 2022
Publication date	Nov 26, 2024
Grant date	Nov 26, 2024

Abstract

A system for guiding interactions with a user device requests a response from first users, stores the response as response data forming a subset of a personal data set of each of the responding first users, and generates a predictive model during training of a machine learning program with a training data set including the personal data set of each of the first users. The predictive model predicts a predicted response of a second user associated with the user device by correlating a personal data set of the second user to the personal data set of at least one of the first users. The computer sends a communication to the user device of the second user having content relating to the first product and/or service when it is determined that the predicted response indicates an interest in or preference of the second user for the first product and/or service.

First claim

We claim:

1. A system for guiding interactions with a user device, the system comprising:
a computer with one or more processor and memory, wherein the computer executes computer-readable instructions to guide the interactions with the user device, the computer managed by a business entity; and
a network connection operatively connecting the user device to the computer;
wherein, upon execution of the computer-readable instructions, the computer performs steps comprising:
requesting a plurality of responses to a plurality of queries from a plurality of first users, wherein each of the first users has an account with the business entity, wherein the account of each of the first users is accessible via a software application and/or a website managed by the computer of the business entity, wherein the account of each of the first users is associated with a respective account setting, wherein the account setting of each account is associated with a manner in which the computer interacts with a corresponding user device of each of the respective first users via the software application and/or the website managed by the computer of the business entity, wherein each of the responses is requested when each of the respective first users is enrolling or participating in a contest, wherein eligibility for a prize of the contest requires each corresponding first user providing responses to the plurality of queries, wherein the plurality of queries includes a first query and a second query, wherein the first query includes a relationship between the contest and a first financial product and/or service offered by the business entity managing the computer, wherein a first response to the first query includes information relating to whether the corresponding first user has an interest in or preference for a first financial product and/or service, and wherein a second response to the second query relates to a preference of each of the respective first users regarding a selection of the account setting for determining the manner in which the computer interacts with the respective user device of each of the respective first users via the software application and/or the website managed by the computer of the business entity;
storing the responses of each of the first users as response data, the response data forming a subset of a personal data set of each of the first users;
training, via an iterative training and testing loop, a machine learning program utilizing at least one neural network to generate a trained predictive model, a training data set utilized during the training of the machine learning program comprising the personal data set of each of the first users, the training comprising:
inserting a target variable value into the iterative training and testing loop;
iteratively predicting the target variable via the iterative training and testing loop, wherein iterative predictions of the target variable comprise modifying weights and calculations applied to the training data set during subsequent prediction iterations in order to improve predictability of the target variable;

deploying the trained predictive model;
predicting, by the predictive model, a predicted first response of a second user associated with the user device to the first query, wherein the second user has an account with the business entity that is accessible via the software application and/or the website managed by the computer of the business entity, wherein the account of the second user is associated with the account setting, wherein a personal data set of the second user, as acquired by the business entity managing the computer, is stored to the memory of the computer, the predicting of the predicted first response of the second user including the predictive model correlating the personal data set of the second user to the personal data set of at least one of the first users, wherein the predicting of the predicted first response of the second user occurs in reaction to a determination by the computer, during a monitoring of changes or additions to the personal data set of the second user, that a triggering condition has been met with respect to at least one entry of the personal data set of the second user;
predicting, by the predictive model, a probability that the second user will use and/or purchase the first financial product and/or service by correlating the personal data set of the second user to the personal data set of at least one of the first users;
predicting, by the predictive model, a predicted second response of the second user to the second query by correlating the personal data set of the second user to the personal data set of at least one of the first users, wherein the predicted second response corresponds to a predicted preference of the second user regarding the selection of the account setting;
automatically selecting the account setting of the second user to correspond to the predicted preference of the second user regarding the selection of the account setting, the automatic selecting of the account setting corresponding to an automatic reconfiguring of the computer to result in a change in one or more of:
a form, a frequency, or a content of communications from the computer that are accessible to the second user during execution of the software application and/or the web site managed by the computer of the business entity on the user device,
a condition triggering the computer to communicate with the second user during execution of the software application and/or the website managed by the computer of the business entity on the user device, and/or
a configuration of an interface of the software application and/or the web site managed by the computer during execution thereof on the user device;

sending, via the network connection, a communication to the user device of the second user when it is determined that the predicted first response indicates an interest in or preference of the second user for the first financial product and/or service and that a threshold value has been met or exceeded regarding the probability that the second user will use and/or purchase the first financial product and/or service, the communication including content relating to the first financial product and/or service, and wherein the communication is displayed to the second user during execution of the software application and/or the website managed by the computer of the business entity on the user device in accordance with the automatically selected account setting of the second user corresponding to the predicted preference of the second user regarding the selection of the account setting.

2. The system of claim 1 , wherein the first query is related to a future use of a prize of the contest by the corresponding first user.

3. The system of claim 2 , wherein the first query is related to the corresponding first user purchasing the first financial product and/or service via use of the prize of the contest.

4. The system of claim 1 , wherein the responses are requested when each of the plurality of first users is navigating the software application and/or the website managed by the computer of the business entity when executed on a corresponding user device.

5. The system of claim 1 , wherein the training of the machine learning program includes unsupervised learning wherein each of the entries of the training data set is unlabeled.

6. The system of claim 1 , wherein the training of the machine learning program includes semi-supervised learning, wherein during the semi-supervised learning the training data set further includes data relating to an evaluation of an accuracy of the predicted first response of the second user and/or an accuracy of the predicted second response of the second user.

7. The system of

Assignees

• Truist Bank

Inventors

• Wright David
• Pham David
• Lewis Adam Thomas

Classifications

• G06Q30/0212 Primary

  Chance discounts or incentives · CPC title

• G06Q30/0224

  based on user history · CPC title

• G06N3/08

  Learning methods · CPC title

• G06N3/0895

  Weakly supervised learning, e.g. semi-supervised or self-supervised learning · CPC title

• G06N3/09

  Supervised learning · CPC title