Using machine learning methods for detecting network anomalies within SNMP-MIB dataset

Naymat, Ghazi Al; Kasassbeh, Mouhammd Al; Harwari, Eshraq Al

doi:10.1504/ijwmc.2018.094644

Cited by 21 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A distributed denial-of-service (DDoS) attack is a self-imposed mechanism to disrupt normal network traffic by flooding the network to the point where the network or associated resources are significantly slowed or crippled. Many studies [13,23,24] have used Simple Network Management Protocol (SNMP-MIB) data. Sometimes, statistical analysis of Management Information Base (MIB) data is also used.…”

Section: Related Workmentioning

confidence: 99%

Cyber Threat Intelligence for IoT Using Machine Learning

2022

View full text Add to dashboard Cite

The Internet of Things (IoT) is a technological revolution that enables human-to-human and machine-to-machine communication for virtual data exchange. The IoT allows us to identify, locate, and access the various things and objects around us using low-cost sensors. The Internet of Things offers many benefits but also raises many issues, especially in terms of privacy and security. Appropriate solutions must be found to these challenges, and privacy and security are top priorities in the IoT. This study identifies possible attacks on different types of networks as well as their countermeasures. This study provides valuable insights to vulnerability researchers and IoT network protection specialists because it teaches them how to avoid problems in real networks by simulating them and developing proactive solutions. IoT anomalies were detected by simulating message queuing telemetry transport (MQTT) over a virtual network. Utilizing DDoS attacks and some machine learning algorithms such as support vector machine (SVM), random forest (RF), k-nearest neighbors (KNN) and logistic regression (LR), as well as an artificial neural network, multilayer perceptron (MLP), naive Bayes (NB) and decision tree (DT) are used to detect and mitigate the attack. The proposed approach uses a dataset of 4998 records and 34 features with 8 classes of network traffic. The classifier RF showed the best performance with 99.94% accuracy. An intrusion detection system using Snort was implemented. The results provided theoretical proof of applicability and feasibility.

show abstract

Section: Related Workmentioning

confidence: 99%