Machine-Learning Based Threat-Aware System in Software Defined Networks

SDN based Intrusion Detection and Prevention Systems using Manufacturer Usage Description: A Survey

“…A study EUNOIA [106] proposes a system to detect the threat and respond in the SDN system. The framework has multiple modules like data preprocessing, decision making, response system, and predictive data modelling.…”

Artificial intelligence enabled software‐defined networking: a comprehensive overview

“…In this context, NNs [97][98][99][100][101][102][103][104][105][106], SVM [105][106][107][108][109][110][111][112][113], DTs [114][115][116][117][118][119][120][121][122][123][124][125][126][127][128], ensemble methods [129][130][131][132][133][134][135][136][137][138][139][140] and supervised deep learning [44,[141][142][143][144] approaches were the most used...…”

“…Stadler et al [131] proposed a RF approach for estimating service-level metrics based on network-level statistics. Song et al [132] presented an intrusion detection system where the RF approach has been used for both feature selection and classification. Miettinen et al [133] introduced, IoT SENTINEL, an SDN-based system for automatic identification and security enforcement of potentially vulnerable IoT devices using device type-specific profiling approach.…”

“…Stadler et al [131] RF estimating service-level metrics outperformed regression tree (RT) in terms of estimation accuracy. However, RF is 3× longer than RT in terms computation time Song et al [132] RF intrusion detection an accuracy of 0.99% on KDD99 Miettinen et al [133] RF automatic identification and security enforcement for IoT devices an accuracy of 0.815% and low execution time (<1 ms)…”

“…In this context, NNs [97–106], SVM [105–113], DTs [114–128], ensemble methods [129–140] and supervised deep learning [44, 141–144] approaches were the most used supervised learning techniques in SDN.…”