US 12,406,143 B2
Systems and methods for identifying and removing bias from communications
Fatima Ahmed, Pflugerville, TX (US); Noor Portillo, Mansfield, TX (US); Aditya Patel, Frisco, TX (US); Sharat Nayak, Dallas, TX (US); Nick Stanton, Bristol, PA (US); David M Lin, Addison, TX (US); and Madhulata Guttikonda, Allen, TX (US)
Assigned to JPMORGAN CHASE BANK, N.A., New York, NY (US)
Filed by JPMORGAN CHASE BANK, N.A., New York, NY (US)
Filed on Oct. 12, 2022, as Appl. No. 18/045,939.
Prior Publication US 2024/0126995 A1, Apr. 18, 2024
Int. Cl. G06F 40/30 (2020.01); G06F 40/279 (2020.01)
CPC G06F 40/30 (2020.01) [G06F 40/279 (2020.01)] 17 Claims
OG exemplary drawing
 
1. A method for identifying bias, comprising:
receiving, by a bias identification computer program executed by an electronic device and from a user electronic device, text comprising a plurality of passages;
converting, by the bias identification computer program, each of the plurality of passages into a vector;
retrieving, by the bias identification computer program, a plurality of custom entities from a custom entity database, wherein the custom entity database comprises a plurality of categories including a plurality of bias categories, wherein each of the bias categories comprises terms and phrases associated with bias;
identifying, by the bias identification computer program, a plurality of custom entities in the vectors;
determining, by the bias identification computer program, that one of the custom entities from one of the plurality of vectors or in a preceding or subsequent vector is in one of the bias categories using a trained bias identification machine learning engine;
returning, by the bias identification computer program and to the user electronic device, an indication that the one of the plurality of vectors is biased;
identifying, by the bias identification computer program, a replacement word for the custom entity that is in the bias category;
replacing, by the bias identification computer program, the vector in the bias category with the replacement word; and
outputting, by the bias identification computer program, the text with the replacement word.