Cybersecurity

Table 7 Illustration of trade-offs between different ROP detection methods

From: Tackling imbalanced data in cybersecurity with transfer learning: a case with ROP payload detection

Method	Detection performance	Overhead
DROP (Chen et al. 2009)	0% FPR	\(\sim\) 500%
kBouncer (Pappas et al. 2013)	0% FPR	\(\sim\) 4%
CFI (Bletsch et al. 2011)	N/A	0–5% majority \(\sim\) 20% worst case
Our method	2.9% Model FPR 95.21% DR	0%

Back to article page