A Feature Selection Method for Large-Scale Network Traffic Classification Based on Spark

Identifying WeChat Message Types without Using Traditional Traffic

Attackers can eavesdrop and exploit user privacy by classifying traffic into different types of in-app service usage to identify user actions. WeChat is the largest social messaging platform, which is a popular application in China. When WeChat is shut down, it is unable to generate traffic; that is, traditional traffic. However, the traffic still can be generated by system. How to identify the message types within WeChat with traffic generated by a system instead of traditional traffic becomes a new challenge. To deal with this challenge, we designed a system to identify and analyze the traffic of the Apple Push Notification service (APNs) to identify the message types of WeChat. In detail, we designed a system to identify and analyze the traffic of the APNs. First, the system clusters the traffic based on the session and divides it into multiple bursts. Then, it extracts the features of each burst and sends these features to the learning-based classifier to extract APNs’s traffic from the background traffic. Finally, it uses a hash-based lookup table method to analyze message types from APNs traffic. Extensive evaluation results show that we can accurately identify the six message types of APN and WeChat. In addition, we propose two coping strategies for the method proposed in this article.

MODC: A Pareto-Optimal Optimization Approach for Network Traffic Classification Based on the Divide and Conquer Strategy

Network traffic classification aims to identify categories of traffic or applications of network packets or flows. It is an area that continues to gain attention by researchers due to the necessity of understanding the composition of network traffics, which changes over time, to ensure the network Quality of Service (QoS). Among the different methods of network traffic classification, the payload-based one (DPI) is the most accurate, but presents some drawbacks, such as the inability of classifying encrypted data, the concerns regarding the users’ privacy, the high computational costs, and ambiguity when multiple signatures might match. For that reason, machine learning methods have been proposed to overcome these issues. This work proposes a Multi-Objective Divide and Conquer (MODC) model for network traffic classification, by combining, into a hybrid model, supervised and unsupervised machine learning algorithms, based on the divide and conquer strategy. Additionally, it is a flexible model since it allows network administrators to choose between a set of parameters (pareto-optimal solutions), led by a multi-objective optimization process, by prioritizing flow or byte accuracies. Our method achieved 94.14% of average flow accuracy for the analyzed dataset, outperforming the six DPI-based tools investigated, including two commercial ones, and other machine learning-based methods.