US 12,271,720 B1
System and methods for detecting required rule engine updates using artificial intelligence models
Miriam Silver, Tel Aviv (IL); and James Myers, New York, NY (US)
Assigned to Citibank, N.A., New York, NY (US)
Filed by Citibank, N.A., New York, NY (US)
Filed on Jul. 23, 2024, as Appl. No. 18/781,965.
Application 18/781,965 is a continuation in part of application No. 18/535,001, filed on Dec. 11, 2023, granted, now 12,045,610.
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 8/60 (2018.01)
CPC G06F 8/60 (2013.01) 18 Claims
OG exemplary drawing
 
1. A system for detecting required rule engine updated using artificial intelligence models that do not require training specific to model components and objectives, the system comprising:
one or more processors; and
one or more non-transitory, computer-readable media comprising instructions recorded thereon that when executed by the one or more processors cause operations comprising:
retrieving historical data for a first rule engine, wherein the historical data comprises historical inputs used by the first rule engine to generate historical outputs, wherein the historical data comprises processing data traversing a cloud computing network, and wherein the first rule engine comprises processing rules for directing processing requests through the cloud computing network;
determining, using a first artificial intelligence model, a first trend in a first subset of the historical inputs, wherein the first artificial intelligence model does not require the historical data for training;
determining a first data characteristic for the first subset based on the first trend;
retrieving a first actual input for the first rule engine, wherein the first actual input is processed by the first rule engine to generate a first actual output, and wherein the first actual input comprises data for a processing request being processed through the cloud computing network;
replacing a first value in a first actual input based on the first data characteristic to generate a first modified input;
inputting the first modified input into the first rule engine, wherein the first rule engine processes the first modified input to generate a first modified output;
determining a first difference between the first modified output and the first actual output, wherein determining the first difference between the first modified output and the first actual output further comprises:
determining a first variable in the first modified output;
determining a second variable in the first actual output; and
determining the first difference based on the first variable and the second variable;
comparing the first difference to a first model deployment criterion; and
determining whether to deploy a first modified rule engine based on comparing the first difference to the first model deployment criterion, wherein the first model deployment criterion corresponds to a requirement for an update to the first rule engine.