US 12,093,872 B2
Evaluating effects of an artificial intelligence model on enterprise performance objectives
Ruchi Mahindru, Elmsford, NY (US); Daniela Rosu, Mount Kisco, NY (US); and Atul Kumar, Bangalore (IN)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Oct. 29, 2021, as Appl. No. 17/514,295.
Prior Publication US 2023/0140553 A1, May 4, 2023
Int. Cl. G06Q 10/06 (2023.01); G06Q 10/0637 (2023.01); G06Q 10/0639 (2023.01)
CPC G06Q 10/06375 (2013.01) [G06Q 10/06393 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
a memory that stores computer executable components; and
a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
a detection component that:
detects, via a machine learning model, a first trend associated with first performance metrics tracked for an enterprise system, wherein the first trend comprises at least one first change to at least one of the first performance metrics, and
correlates, via the machine learning model, the first trend with a second trend associated with second performance metrics tracked for an artificial intelligence model employed by the enterprise system that performs an automated task for the enterprise system, wherein the automated task comprises a customer service interaction task, wherein the second trend comprises at least one second change to at least one of the second performance metrics,
wherein the first performance metrics measure one or more business objectives of an enterprise associated with the enterprise system, and the first performance metrics are not determined directly using any output of the artificial intelligence model, and
wherein the second performance metrics measure one or more performance objectives of the artificial intelligence model, and the second performance metrics are determined directly using one or more outputs of the artificial intelligence model;
a correlation component that determines, via the machine learning model, based on the correlation of the first trend and the second trend, a technical issue associated with the artificial intelligence model that correlates to the at least one first change to the at least one of the first performance metrics using a data model that defines first relationships between the second performance metrics and candidate technical issues associated with the artificial intelligence model;
a remediation component that determines, via the machine learning model, a solution for the technical issue; and
a model updating component that updates, via the machine learning model, the artificial intelligence model according to a configuration change based on a determination that the solution comprises the configuration change, resulting in an updated artificial intelligence model.