Redirigiendo al acceso original de articulo en 17 segundos...
Inicio  /  Future Internet  /  Vol: 15 Par: 12 (2023)  /  Artículo
ARTÍCULO
TITULO

Securing Network Traffic Classification Models against Adversarial Examples Using Derived Variables

James Msughter Adeke    
Guangjie Liu    
Junjie Zhao    
Nannan Wu and Hafsat Muhammad Bashir    

Resumen

Machine learning (ML) models are essential to securing communication networks. However, these models are vulnerable to adversarial examples (AEs), in which malicious inputs are modified by adversaries to produce the desired output. Adversarial training is an effective defense method against such attacks but relies on access to a substantial number of AEs, a prerequisite that entails significant computational resources and the inherent limitation of poor performance on clean data. To address these problems, this study proposes a novel approach to improve the robustness of ML-based network traffic classification models by integrating derived variables (DVars) into training. Unlike adversarial training, our approach focuses on enhancing training using DVars, introducing randomness into the input data. DVars are generated from the baseline dataset and significantly improve the resilience of the model to AEs. To evaluate the effectiveness of DVars, experiments were conducted using the CSE-CIC-IDS2018 dataset and three state-of-the-art ML-based models: decision tree (DT), random forest (RF), and k-neighbors (KNN). The results show that DVars can improve the accuracy of KNN under attack from 0.45% to 0.84% for low-intensity attacks and from 0.32% to 0.66% for high-intensity attacks. Furthermore, both DT and RF achieve a significant increase in accuracy when subjected to attack of different intensity. Moreover, DVars are computationally efficient, scalable, and do not require access to AEs.

 Artículos similares

       
 
Rasheed Ahmad, Izzat Alsmadi, Wasim Alhamdani and Lo?ai Tawalbeh    
Today, deep learning approaches are widely used to build Intrusion Detection Systems for securing IoT environments. However, the models? hidden and complex nature raises various concerns, such as trusting the model output and understanding why the model ... ver más
Revista: Future Internet

 
Louiza Hamada, Pascal Lorenz and Marc Gilg    
Although visible light communication (VLC) channels are more secure than radio frequency channels, the broadcast nature of VLC links renders them open to eavesdropping. As a result, VLC networks must provide security in order to safeguard the user?s data... ver más
Revista: Future Internet

 
Blerim Rexha, Gresa Shala and Valon Xhafa    
Personal mobile devices currently have access to a significant portion of their user?s private sensitive data and are increasingly used for processing mobile payments. Consequently, securing access to these mobile devices is a requirement for securing ac... ver más
Revista: Future Internet

 
Jin-Yong Lee, Kideok D. Kwon     Pág. 1 - 16
Korea has been operating groundwater monitoring systems since 1996 as the Groundwater Act enacted in 1994 enforces nationwide monitoring. Currently, there are six main groundwater monitoring networks operated by different government ministries with diffe... ver más
Revista: Water

 
Niklas Galonske, Eckhard Riebe, Armand Toubol, Sebastian Weismantel     Pág. 2850 - 2859
Rail transports of single wagons or wagon groups are an indispensable part of the transport chain, such as for the forestry and the chemical industry. However, high production costs and low quality standards have led to a continuous decline in market sha... ver más