US 12,265,502 B1
Multi-program applicant review system with adjustable parameters
Alexander Thomson, Philadelphia, PA (US); Tanner Mitton, Irving, TX (US); and David Elder, New York, NY (US)
Assigned to MEDICRATIC INC., Irving, TX (US)
Filed by Medicratic Inc., Richardson, TX (US)
Filed on Mar. 4, 2024, as Appl. No. 18/594,912.
Claims priority of provisional application 63/450,827, filed on Mar. 8, 2023.
Int. Cl. G06F 16/16 (2019.01)
CPC G06F 16/164 (2019.01) 18 Claims
OG exemplary drawing
 
1. A method to assess applicant data, the method comprising:
receiving, using one or more processors of a computing device, input applicant data;
receiving, using the one or more processors and one or more user interfaces of the computing device, one or more inputs establishing one or more preference parameters for an applicant assessment platform;
performing, with the one or more processors of the computing device, a machine-learning (ML)-based assessment of the input applicant data to generate a ML-based score corresponding to the input applicant data and the one or more preference parameters, the ML-based assessment including at least:
a normalization of text data from the input applicant data to create one or more normalized files,
a semantic embedding of the one or more normalized files to form one or more embedded normalized files, and
a bias mitigation operation including:
detecting one or more potentially biasing identifiers from one or more document files, and
replacing the one or more potentially biasing identifiers with one or more corresponding neutral terms such that a semantic structure is maintained; and
presenting, at the one or more user interfaces of the computing device, an applicant ranking based at least partly on the ML-based score.