Random Forest, Logistic Regression, and Naive Bayes models for intrusion detection reach 95-99% accuracy on single datasets but fall below 40% in cross-dataset tests between UNSW-NB15 and TON_IoT.
Mesopotamian Journal of CyberSecurity2021, 1–4 (2021)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Assessing Generalisation Capability of Machine Learning Models for Intrusion Detection
Random Forest, Logistic Regression, and Naive Bayes models for intrusion detection reach 95-99% accuracy on single datasets but fall below 40% in cross-dataset tests between UNSW-NB15 and TON_IoT.