US 12,112,370 B2
Secure multi-party computation for sensitive credit score computation
Fardin Abdi Taghi Abad, Champaign, IL (US); Austin Walters, Savoy, IL (US); Jeremy Goodsitt, Champaign, IL (US); Anh Truong, Champaign, IL (US); Mark Watson, Urbana, IL (US); and Reza Farivar, Champaign, IL (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Sep. 1, 2022, as Appl. No. 17/929,229.
Application 17/929,229 is a continuation of application No. 16/893,557, filed on Jun. 5, 2020, granted, now 11,436,671.
Prior Publication US 2022/0414768 A1, Dec. 29, 2022
Int. Cl. G06Q 40/03 (2023.01); H04L 9/08 (2006.01)
CPC G06Q 40/03 (2023.01) [H04L 9/0861 (2013.01); G06Q 2220/00 (2013.01); H04L 2209/46 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
retrieving, by a computing system, an algorithm, the algorithm utilizing one or more parameters for evaluation of a user metric associated with an individual;
identifying, by the computing system, a plurality of entities contributing parameters for the evaluation of the user metric;
compiling, by the computing system, the algorithm into a plurality of components, wherein each component corresponds to a respective entity of the plurality of entities and each component generates an output unique to the respective entity, the compiling comprising:
converting the algorithm into one or more Boolean circuit representations, and
garbling each Boolean circuit representation, and
executing a protocol across the plurality of entities for securely evaluating the garbled Boolean circuit representations, wherein the protocol prevents exposure of personal identification information of the individual with among the plurality of entities;
transmitting, by the computing system, each component to a respective entity of the plurality of entities;
receiving, by the computing system, an evaluated component from each entity of the plurality of entities;
verifying, by the computing system, that the received evaluated components are matching,
wherein if at least one evaluated component does not match, the computing system reinitiates the executing, transmitting, and receiving steps until the evaluated components are matching; and
generating, by the computing system upon verifying that the evaluated components are matching, the user metric associated with the individual by combining the evaluated components from each entity.