Efficient Network Traffic Classification and Visualizing Abnormal Part Via Hybrid Deep Learning Approach : Xception + Bidirectional GRU

Due to a rapid development in the field of information and communication, technologies yielded novel changes both individual organizational operations. Therefore, accessibility became easier more convenient than before, malicious approaches such as hacking or spying aimed at various kept increasing. With aim preventing approaches, classification detecting traffic are vital. our research utilized deep learning machine models for better classification. The given dataset consists normal data these types png files. In order achieve precise classification, experiment three steps. Firstly, only vanilla CNN was used highest score 86.2%. Second all, hybrid approach, classifiers were instead fully connected layers from it about 87% with extra tree classifier.At last, Xception model combined bidirectional GRU attained 95.6% accuracy score, which among all.