Search for potentially dangerous constructs in the source code of the program using neural networks
2021 ◽
Vol 2131
(2)
◽
pp. 022111
Keyword(s):
Abstract This work is devoted to the study of artificial neural networks to search for potentially dangerous constructs in the source code of the program. A modification of the existing analyzer programs is proposed for a better search for vulnerabilities in software. The use of neural networks will allow programmers and analysts to find those vulnerabilities that have not yet been added to the database. As a result of the research, it was found that the use of neural networks to identify specified constructs is a more effective solution than searching through a database of vulnerabilities, since it allows you to identify not only those constructs that are in the database, but also suspicious code sections.
1999 ◽
Vol 22
(8)
◽
pp. 723-728
◽
2002 ◽
Vol 20
(2)
◽
pp. 89-95
◽
Keyword(s):
2005 ◽
Vol 53
(5)
◽
pp. 343
◽
Keyword(s):
Keyword(s):
2012 ◽
Vol 3
(2)
◽
pp. 48-50
Keyword(s):