| CPC G06Q 10/0635 (2013.01) [G06F 18/2193 (2023.01); G06F 18/2415 (2023.01); G06N 20/00 (2019.01); G06Q 10/083 (2013.01)] | 20 Claims |

|
1. A computer-implemented method comprising:
receiving, by a processor, a plurality of policy parameters associated with a provider of a plurality of providers from a plurality of pre-generated databases;
dynamically enriching, by the processor, input data associated with the plurality of policy parameters and a plurality of identified data records,
wherein enriched input data is utilized to train a machine learning model associated with the identified data record;
training, by the processor, a machine learning model with at least one data feedback loop by introducing training data as a plurality of variables to generate a first plurality of scenarios in real-time,
dynamically simulating, by the processor, the first plurality of scenarios in real time to optimize a first respective dynamic probability risk value generated by a respective dynamic data model using the trained machine learning model;
automatically updating, by the processor, the at least one feedback loop associated with the trained machine learning model;
dynamically generating, by the processor, a second plurality of scenarios in real-time based on the at least one feedback loop associated with the trained machine learning model;
dynamically determining, by the processor, a predetermined policy risk threshold for the identified data record based on the second plurality of scenarios; and
automatically modifying, by the processor, the predetermined policy risk threshold associated with a qualified provider of the plurality of providers based on a respective model risk probability value to generate a modified policy risk threshold associated with the identified data record.
|