CPC G06F 11/079 (2013.01) [G06F 11/0709 (2013.01); G06F 11/0793 (2013.01)] | 21 Claims |
1. A method comprising:
receiving, by one or more processors, new incident data associated with a new incident of downtime or an interrupted service of one or more of a first computing application, a first computing device, or a first computing service; and
determining, by the one or more processors, by inputting the new incident data into a trained machine learning model, that the new incident is related to a first change to the first computing application, the first computing device, or the first computing service, wherein:
the trained machine learning model has been trained using change causation data indicating whether each of a plurality of historical incidents of downtime or interrupted service of at least one second computing application, at least one second computing device, or at least one second computing service is related to one or more second changes to the at least one second computing application, the at least one second computing device, or the at least one second computing service; and
training of the trained machine learning model is further based on incident data associated with each of the plurality of historical incidents, wherein the incident data, for each of the plurality of the historical incidents, comprises version history information indicating one or more changes to the at least one second computing application, the at least one second computing device, or the at least one second computing service.
|