US 11,853,913 B1
Systems and methods for dynamic adjustment of computer models
Xiaomin Lin, Springfield, MA (US); Matthew Girard, Springfield, MA (US); Michael Crough, Springfield, MA (US); Peng Wang, Springfield, MA (US); Adam Fox, Springfield, MA (US); and Robert Greif, Springfield, MA (US)
Assigned to Massachusetts Mutual Life Insurance Company, Springfield, MA (US)
Filed by Massachusetts Mutual Life Insurance Company, Springfield, MA (US)
Filed on Jan. 3, 2023, as Appl. No. 18/092,682.
Application 18/092,682 is a continuation of application No. 16/906,423, filed on Jun. 19, 2020, granted, now 11,544,598.
Claims priority of provisional application 62/863,730, filed on Jun. 19, 2019.
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 15/16 (2006.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G06N 5/04 (2013.01) [G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
executing, by a server, an artificial intelligence model to identify a first weight factor used by an artificial intelligence model to generate a first score for a user,
the artificial intelligence model having been previously trained based on a training dataset comprising known training user attributes and their corresponding scores,
wherein the artificial intelligence model predicts the first score for the user by applying at least one user attribute and the first weight factor to a dynamic algorithm;
executing, by the server, a computer model using at least one user attribute where the computer model determines a second score for the user using a second weight factor;
prompting, by the server, a confirmation of the second score;
in response to receiving a negative response, identifying, by the server, a third weight factor by retraining the artificial intelligence model, whereby the server generates a third score calculated based on the third weight factor.