Table 1.
Comparative analysis of intelligent methods in software-defined networking (SDN) research.
Fig 1.
Structure of the Gauss Markov and Flow-balanced Vector Radial Learning (GM-FVRL) method.
Fig 2.
Architecture of the network traffic classification method on IoT with SDN.
Fig 3.
Structure of Gauss Markov Correlation-based Feature Extraction model.
Table 2.
Dataset description.
Table 3.
Latency comparison of proposed GM-FVRL versus existing IPro [1], MACCA2-RF&RF [2], and novel hybrid DL model based on CNN [10].
Fig 4.
Classification accuracy comparison of proposed GM-FVRL versus existing IPro [1], MACCA2-RF&RF [2], and novel hybrid DL model based on CNN [10].
Table 4.
Precision comparison of proposed GM-FVRL versus existing IPro [1], MACCA2-RF&RF [2], and novel hybrid DL model based on CNN [10].
Fig 5.
Recall comparison of proposed GM-FVRL versus existing IPro [1], MACCA2-RF&RF [2], and novel hybrid DL model based on CNN [10].
Table 5.
Error rate comparison of proposed GM-FVRL versus existing IPro [1], MACCA2-RF&RF [2], and novel hybrid DL model based on CNN [10].
Fig 6.
F1-measure comparison of proposed GM-FVRL versus existing IPro [1], MACCA2-RF&RF [2], and novel hybrid DL model based on CNN [10].
Table 6.
Comparison of all performance metrics for proposed GM-FVRL versus existing IPro [1], MACCA2-RF&RF [2], and novel hybrid DL model based on CNN [10].