… network traffic classification. In Sect. 3, we present the essential background on deep learning
… Finally, the automatic feature extraction procedure from network traffic can save the cost of …

… classifiers, some of which are manual attribute selection and others are automatic attribute
selection, and we also compare these experimental results with existing works. For the …

… traffic surveys focus on classification and do not consider anonymization. Here, we review
the Internet traffic classification … different objectives of Internet traffic classification. Finally, we …

… classification tasks, including data collection and cleaning, feature selection, and model
selection. Moreover, we discuss deep learning … representation of the auto-encoder and achieve …

… ability of automatic feature learning, Deep Learning (DL) … mobile services traffic classification,
especially encrypted traffic. … the application of DL methods to encrypted traffic classification …

… , traffic classification is also the central part of the automated … Internet traffic classification
using machine learning methods … includes machine learning for network traffic classification. A …

… of (effective) deep packet inspection approaches. Also, the fast-expanding set of apps and
the moving-target nature of mobile traffic makes design solutions with usual machine learning, …

… Internet traffic classification and application identification by … , and then using known image
classification deep learning … “Automated traffic classification and application identification …

… For classification of packets, we use machine learning algorithms in the WEKA tool [17].
WEKA provides a variety of machine learning libraries and we used DecisionTable and J48 …

… Section 3 presents our approach to performing the traffic classification, including both data …
identification and traffic characterisation. An automatic deep learning-based feature extraction …