Application layer classification of Internet traffic using ensemble learning models

Abstract

Summary Accurate application layer classification of Internet traffic has been a necessary requirement for various regulatory, control, and operational purposes of Internet service provider (ISP). Due to the dynamic and ever evolving nature of Internet applications generating a diverse mixture of Internet traffic, it has been necessary to apply deep packet inspection (DPI) techniques for traffic classification. DPI methods offer accuracy but degrade overall network throughput and thus cause problems in ensuring quality of service (QoS) and maintaining service‐level agreements. Moreover, Internet traffic is mostly end to end encrypted. This in turn limits the applicability of DPI techniques and renders them useless, unless the encryption tunnel is broken by the service provider which would risk violating user privacy. To address these trade‐offs between classification accuracy, performance, and user privacy, we resort to machine learning (ML)‐based algorithms. In this article, we apply three ensemble ML algorithms and report their performance metrics in the application layer classification of Internet traffic.

Method for Multi-Task Learning Fusion Network Traffic Classification to Address Small Sample Labels

“…While delving into the sphere of network traffic categorization, bolstered by machine learning and deep learning, academics predominantly apply methodologies such as supervised learning [4] [5], unsupervised learning [6] [7], and collective learning approaches. [8] [9] to distinguish the types of business traffic, such as email, web browsing, video streaming, etc.…”

Identifying Consumer Drones via Encrypted Traffic

“…Recommended model types include convolutional neural networks (CNN), autoencoders (AE), and generative adversarial networks (GAN). These models can be used simultaneously, and the results of each model can be aggregated in a process called ensemble learning [19]. DL methods do not rely on human-extracted features, which may be of convenience, but it is then difficult to justify the impact and reasoning logistically and holistically behind each.…”

“…To further illustrate the generation of the feature vector, we will walk through the following example. We begin by initializing the drone class profiles, each composed of UNV message types: (19), V (26), V (43), V (36),V(54),V (48), V (21),V (23),V(34)} = {0.35, 0.35, 0.19, 0.00, 0.00, 0.00, 0.00, 0.02, 0.03}.…”

Method for multi-task learning fusion network traffic classification to address small sample labels

“…While delving into the sphere of network traffic categorization, bolstered by machine learning and deep learning, academics predominantly apply methodologies such as supervised learning 4 , 5 , unsupervised learning 6 , 7 , and collective learning approaches. 8 , 9 to distinguish the types of business traffic, such as email, web browsing, video streaming, etc.…”