METHOD AND SYSTEM FOR ASSESSING THE PROBABILITY OF CRITICAL CYBER SECURITY DEFECTS AT ACCEPTANCE TESTS OF PRODUCT RELEASES Russian patent published in 2021 - IPC G06N20/10 G06N20/20 G06F17/18 

FIELD: computing.

SUBSTANCE: invention relates to computing. Disclosed is a computer-implemented method for assessing the likelihood of critical cybersecurity defects at acceptance testing of product releases, in which data is obtained containing information about software releases that are in development statuses preceding acceptance testing; processing the received data using a machine learning model, during which they perform: dividing the imported data into categorical and numerical variables; transformation of the obtained variables, performing vectorization of categorical variables and normalization of numerical variables; concatenation of the transformed variables and building on their basis a vector corresponding to the estimated releases in the statuses preceding the acceptance tests; classification using the resulting vector of each release, assigning the degree of probability of critical cybersecurity defects at acceptance tests of product releases.

EFFECT: increased speed and accuracy of assessing the likelihood of critical cybersecurity defects during acceptance tests of product releases.

7 cl, 4 dwg