US 11,755,950 B2
Methods for refining data set to represent output of an artificial intelligence model
Prem Piyush Goyal, Hyderabad (IN); Manish Anand Bhide, Hyderabad (IN); Harivansh Kumar, Hyderabad (IN); and Venkata R. Madugundu, Hyderabad (IN)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Jun. 30, 2020, as Appl. No. 16/916,793.
Prior Publication US 2021/0406762 A1, Dec. 30, 2021
Int. Cl. G06N 20/00 (2019.01); G06F 18/214 (2023.01); G06F 18/2451 (2023.01); G06Q 10/067 (2023.01); G06F 18/24 (2023.01); G06Q 40/12 (2023.01)
CPC G06N 20/00 (2019.01) [G06F 18/214 (2023.01); G06F 18/24 (2023.01); G06F 18/2451 (2023.01); G06Q 10/067 (2013.01); G06Q 40/12 (2013.12)] 18 Claims
OG exemplary drawing
 
1. A method comprising:
obtaining, by a computing system, a dataset used to interpret a binary decision of an artificial intelligence model, the dataset being used to make the binary decision;
generating, by the computing system, one or more constraints for the obtained dataset;
identifying, by the computing system, a first data point for identification of one or more insights thereabout;
generating, by the computing system, a plurality of data points from the dataset that are in proximity to the first data point;
identifying, by the computing system, a subset of data points from the generated plurality of data points satisfying the one or more constraints;
applying, by the computing system, a linear model on the identified subset of data points satisfying the one or more constraints; and
generating, by the computing system and based on the linear model, the one or more insights illustrating the binary decision of the artificial intelligence model.