US 11,790,256 B2
Analyzing test result failures using artificial intelligence models
Lukasz G Cmielowski, Cracow (PL); Maksymilian Erazmus, Cracow (PL); Rafal Bigaj, Cracow (PL); and Wojciech Sobala, Cracow (PL)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Jul. 22, 2022, as Appl. No. 17/871,642.
Application 17/871,642 is a continuation of application No. 16/815,952, filed on Mar. 11, 2020, granted, now 11,475,326.
Prior Publication US 2022/0358381 A1, Nov. 10, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 5/04 (2023.01); G06N 20/00 (2019.01); G06F 11/36 (2006.01)
CPC G06N 5/04 (2013.01) [G06F 11/3692 (2013.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for analyzing test result failures using artificial intelligence models, the method comprising:
clustering failed tests in one or more sets of test failure groups into a set of clusters;
identifying a root cause failure for each cluster in said set of clusters;
training a first machine learning model to predict a root cause of an unclassified failure based on identifying said root cause failure for each cluster in said set of clusters; and
predicting said root cause of said unclassified failure using said first machine learning model.