Optimal Tree Bayesian for the Characterization of Ciphered Network Communication Traffic

초록

The increasing variability of ciphered network communication has elevated the difficulty and complexity of network management. Finding the best optimal solution for identifying and categorizing network traffic requires a methodical approach. To predict heterogeneous ciphered network traffic, this study presented a Tree Bayesian Optimization (TBO) technique. The proposed TBO outperformed others in characterizing diverse, complicated encoded network traffic. The experimental results on the ISCXVPN2016 datasets demonstrate the utility and practicality of the proposed optimization approach, which surpasses current optimization approaches utilized for unassailable network stream evaluation.