US 12,475,381 B2
Systems and methods for personalized guidance
Rohan Vimal Raj, Plano, TX (US); Cole Coen, Celina, TX (US); Krishna C. Chinthakayala, Frisco, TX (US); and Humberto Hernandez Torres, Plano, TX (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Mar. 13, 2024, as Appl. No. 18/603,686.
Claims priority of provisional application 63/490,131, filed on Mar. 14, 2023.
Prior Publication US 2024/0311653 A1, Sep. 19, 2024
Int. Cl. G06F 7/00 (2006.01); G06N 5/01 (2023.01)
CPC G06N 5/01 (2023.01) 20 Claims
OG exemplary drawing
 
1. A method for personalized search results, comprising:
receiving, by one or more processors, a first data object, the first data object including one or more sequentially ordered user inputs;
receiving, by the one or more processors, a second data object, the second data object including data associated with a plurality of items;
determining a plurality of filters based on the first data object,
wherein each filter includes a priority indicator, and
wherein each of the plurality of filters is based on at least part of the one or more user inputs;
determining a resulting number of items in the second data object based on applying the plurality of filters;
determining a pre-defined threshold has not been met by the resulting number of items;
iteratively removing a filter having a lowest priority indicator of the plurality of filters until the resulting number of items satisfies the pre-defined threshold;
generating a token for each one or more user input, wherein classifications associated with one or more removed filters are excluded from the token;
inputting the one or more generated tokens in sequential combination to a machine-learning model;
generating a third data object based on an output of the machine-learning model; and
displaying the third data object to a user.