US 11,954,693 B2
Dynamic test suite creation using event communications from customers
Paola Martinez Morales, Wendell, NC (US); Eric R. Kern, Chapel Hill, NC (US); Robert Furda, Bratislava (SK); Asmaa El Andaloussi, Asnieres-sur-Seine (FR); Firoz Rangwalla, Suwanee, GA (US); and Brian E. Finley, Allen, TX (US)
Assigned to Lenovo Global Technology (United States) Inc., Morrisville, NC (US)
Filed by Lenovo Global Technology (United States) Inc., Morrisville, NC (US)
Filed on Mar. 3, 2022, as Appl. No. 17/686,228.
Prior Publication US 2023/0281637 A1, Sep. 7, 2023
Int. Cl. G06Q 30/00 (2023.01); G06N 20/00 (2019.01); G06Q 30/016 (2023.01)
CPC G06Q 30/016 (2013.01) [G06N 20/00 (2019.01)] 15 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, via a management system, an event communication from a customer about an adverse event, the customer receiving support for computing equipment over a management network from a support provider, the adverse event regarding one or more computing devices of the computing equipment, the management network comprising a computer network connected to computing devices of the computer equipment via a management controller in each of the computing devices, wherein the support provider communicates with the computing devices over the management network;
analyzing the event communication received from the customer using natural language processing to identify a potential cause of the adverse event, the potential cause related to the computing equipment;
selecting one or more tests from a test library based on the identified potential cause of the adverse event, each test of the one or more tests is configured to test a portion of the computing equipment to lead to identification of a cause of the adverse event;
automatically initiating the selected one or more tests through the management network by transmitting code comprising the one or more tests via the management controller, the code executable by the computing devices and/or the management controllers;
analyzing test results from execution of the selected one or more tests to identify a cause the adverse event;
using a machine learning algorithm to correlate tests from the test library with adverse events from event communications; and
training the machine learning algorithm using:
a plurality of event communications from one or more customers about adverse events;
related identified potential causes of the adverse events from the event communications;
configuration information about computing equipment of the one or more customers;
selected tests for each identified potential cause;
analysis from the test results;
identified causes of the adverse events; and/or
confirmed causes of the adverse events.