Automatic data-driven optimization of a target outcome using machine learning

Patent metadata

Field	Value
Publication number	US-12169870-B2
Application number	US-202318108715-A
Country	US
Kind code	B2
Filing date	Feb 13, 2023
Priority date	Feb 13, 2023
Publication date	Dec 17, 2024
Grant date	Dec 17, 2024

Abstract

Methods and apparatuses are described for automatic data-driven optimization of a target outcome using machine learning. A server generates a first feature dataset and applies a trained outcome prediction model to the first feature dataset as input to generate a second feature data set and a first predicted value for a target outcome. The server displays the first predicted value on a client device. The server receives input corresponding to one or more preferences or constraints from the client device and adjusts the trained outcome prediction model based upon the received input to incorporate the one or more preferences or constraints. The server applies the adjusted outcome prediction to the second feature dataset as input to generate a third feature data set a second predicted value for the target outcome. The server displays the second predicted value on the client device.

First claim

What is claimed is:

1. A system for automatic data-driven optimization of a retirement plan target outcome using machine learning, the system comprising a server computing device with a memory for storing computer-executable instructions and a processor that executes the computer-executable instructions to:
generate a user feature dataset comprising static attributes and dynamic attributes for each of a plurality of users;
train an outcome prediction model using the user feature dataset to generate predicted values for a retirement plan target outcome;
identify one or more first user attributes associated with a user of a client computing device;
execute the trained outcome prediction model using the first user attributes to generate one or more second user attributes and a first predicted retirement plan score based upon the second user attributes;
display the second user attributes and the first predicted retirement plan score to the user of the client computing device;
receive user input corresponding to one or more preferences or constraints from the user of the client computing device;
adjust the trained outcome prediction model to incorporate the one or more preferences or constraints by changing one or more of the second user attributes based upon the user input;
execute the adjusted outcome prediction model to generate one or more third user attributes and a second predicted retirement plan score based upon the one or more third user attributes; and
display the third user attributes and the second predicted retirement plan score to the user of the client computing device.

2. The system of claim 1 , wherein identifying one or more first user attributes associated with a user of a client computing device comprises:
receiving a request to access a retirement plan application from the client computing device, the request including authentication credentials associated with the user;
locating a user profile data structure for the user based upon the authentication credentials; and
selecting the one or more first user attributes associated with the user from the user profile data structure.

3. The system of claim 2 , wherein the one or more first user attributes comprise static user attributes and dynamic user attributes.

4. The system of claim 3 , wherein the static user attributes comprise an age of the user, an income of the user and an account balance of the user, and the dynamic user attributes comprise a retirement expense amount of the user and a retirement age of the user.

5. The system of claim 1 , wherein the user input corresponding to one or more preferences or constraints comprises a text string corresponding to an utterance of the user.

6. The system of claim 5 , wherein adjusting the trained outcome prediction model to incorporate the one or more preferences or constraints by changing one or more of the second user attributes based upon the user input comprises:
converting the text string into one or more adjustment operations; and
applying the adjustment operations to change one or more of the second user attributes.

7. The system of claim 6 , wherein converting the text string into one or more adjustment operations comprises:
determining one or more intents associated with the text string using a natural language processor; and
mapping the one or more intents to the one or more adjustment operations.

8. The system of claim 1 , wherein the server computing device:
receives additional user input corresponding to one or more additional preferences or constraints from the user of the client computing device;
re-adjusts the adjusted outcome prediction model to incorporate the one or more additional preferences or constraints by changing one or more of the third user attributes based upon the additional user input;
executes the re-adjusted outcome prediction model to generate one or more fourth user attributes and a third predicted retirement plan score based upon the one or more fourth user attributes; and
displays the fourth user attributes and the third predicted retirement plan score to the user of the client computing device.

9. The system of claim 1 , wherein the server computing device transmits instructions comprising the third user attributes to a remote computing device for updating the user's retirement plan.

10. The system of claim 9 , wherein the server computing device transmits the instructions to the remote computing device upon determining that the second predicted retirement plan score meets or exceeds a predetermined threshold.

11. The system of claim 9 , wherein the server computing device transmits the instructions to the remote computing device upon receiving score acceptance indicia from the user of the client computing device.

12. The system of claim 1 , wherein the outcome prediction model comprises a predictor function and an optimizer function.

13. The system of claim 12 , wherein the predictor function comprises a K-nearest neighbor regression algorithm and the optimizer function comprises a black-box optimization (BBO) algorithm.

14. The system of claim 1 , wherein the user feature dataset comprises a synthetic user feature dataset created through automated variation of the static attributes and the dynamic attributes.

15. A computerized method of automatic data-driven optimization of a retirement plan target outcome using machine learning, the method comprising:
generating, by a server computing device, a user feature dataset comprising static attributes and dynamic attributes for each of a plurality of users;
training, by the server computing device, an outcome prediction model using the user feature dataset to generate predicted values for a retirement plan target outcome;
identifying, by the server computing device, one or more first user attributes associated with a user of a client computing device;
executing, by the server computing device, the trained outcome prediction model using the first user attributes to generate one or more second user attributes and a first predicted retirement plan score based upon the second user attributes;
displaying, by the server computing device, the second user attributes and the first predicted retirement plan score to the user of the client computing device;
receiving, by the server computing device, user input corresponding to one or more preferences or constraints from the user of the client computing device;
adjusting, by the server computing device, the trained outcome prediction model to incorporate the one or more preferences or constraints by changing one or more of the second user attributes based upon the user input;
executing, by the server computing device, the adjusted outcome prediction model to generate one or more third user attributes and a second predicted retirement plan score based upon the one or more third user attributes; and
displaying, by the computing device, the third user attributes and the second predicted retirement plan score to the user of the client computing device.

16. The method of claim 15 , wherein identifying one or more first user attributes associated with a user of a client computing device comprises:
receiving a request to access a retirement plan application from the client computing device, the request including authentication credentials associated with the user;
locating a user profile data structure for the user based upon the authentication credentials; and
selecting the one or more first user attributes associated with the user from the user profile data structure.

17. The method of claim 16 , wherein the

Assignees

• Fmr Llc

Inventors

• Kilitcioglu Doruk
• Kadioglu Serdar

Classifications

• G06Q40/04 Primary

  Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange · CPC title