Methods and apparatus for detecting and preventing obfuscated cyberattacks using machine learning techniques
US11620379B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Sep 25, 2019 |
| Grant date | Apr 4, 2023 |
| Priority date | — |
| Expiry date | Jan 19, 2042 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N20/20
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
The presently disclosed subject matter includes a system for monitoring a set of command lines or calls to executable scripts configured to be executed by an operating system. Each command line from the set of command lines is associated with an executable script configured to be executed by an operating system. The apparatus classifies, via a machine learning model, a command line from the set of command lines into an obfuscation category and prevents the operating system from executing the command line and generates a notification signal when the obfuscation category indicates that the command line is part of a cybersecurity attack. The apparatus allows the operating system to execute the command line or call to the executable script when the obfuscation category indicates that the command line is not part of a cybersecurity attack.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.