| CPC G06F 21/577 (2013.01) [G06N 20/00 (2019.01); G06F 2221/034 (2013.01)] | 20 Claims |
|
1. An apparatus, comprising:
a processor; and
a memory configured to store code executable by the processor to:
automatically identify and label, with a respective component node label, one or more graph component nodes of a data graph representing a computing system under test, wherein each labeled graph component node includes one or more component node security vulnerabilities for the computing system under test;
automatically identify and label, with a respective communication link label, one or more graph communication links between graph component nodes of the data graph, wherein each labeled graph communication link includes one or more communication link security vulnerabilities between graph component nodes for the computing system under test;
utilize the one or more automatically identified and labeled graph component nodes and the one or more automatically identified and labeled graph communication links of the data graph to train a machine learning algorithm to predict one or more misconfigurations in the computing system under test based on the one of the one or more component node security vulnerabilities included in each component node label for the one or more graph component nodes and the one or more communication link security vulnerabilities included in each communication link label for the one or more graph communication links; and
determine one or more modifications to the computing system under test for mitigating the one or more misconfigurations predicted by the machine learning algorithm.
|