Al-Farg'oniy avlodlari, 2026, № 1, с. 356-363
UNSUPERVISED ANOMALY DETECTION IN NETWORK TRAFFIC USING AUTOENCODERS: AN EMPIRICAL STUDY ON THE NSL-KDD DATASET
Аннотация · Abstract
The scarcity of labeled data remains a fundamental limitation of supervised learning models. This study proposes an autoencoder-based unsupervised approach for detecting anomalies in network traffic. Experiments conducted on the NSL-KDD dataset demonstrate that the model, trained exclusively on normal traffic, achieved an AUC-ROC of 96.13% and Precision of 96.25%. The detection of 40.22% of zero-day attacks without any signatures confirms the superiority of autoencoders over traditional signature-based systems. A hybrid approach is recommended as the most effective strategy.
Сведения · Details
- Журнал · Journal Al-Farg'oniy avlodlari
- Язык · Language английский
- Цитирования (Crossref) · Citations 0