Scalable kernel convex hull online support vector machine for intelligent network traffic classification
- PDF / 1,040,921 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 73 Downloads / 248 Views
Scalable kernel convex hull online support vector machine for intelligent network traffic classification Xiaoqing Gu 1 & Tongguang Ni 1 & Yiqing Fan 2 & Weibo Wang 3 Received: 10 June 2019 / Accepted: 22 April 2020 / Published online: 18 June 2020 # Institut Mines-Télécom and Springer Nature Switzerland AG 2020
Abstract Online support vector machine (SVM) is an effective learning method in real-time network traffic classification tasks. However, due to its geometric characteristics, the traditional online SVMs are sensitive to noise and class imbalance. In this paper, a scalable kernel convex hull online SVM called SKCHO-SVM is proposed to solve this problem. SKCHO-SVM involves two stages: (1) offline leaning stage, in which the noise points are deleted and initial pin-SVM classifier is built; (2) online updating stage, in which the classifier is updated with newly arrived data points, while carrying out the classification task. The noise deleting strategy and pinball loss function ensure SKCHO-SVM insensitive to noise data flows. Based on the scalable kernel convex hull, a small amount of convex hull vertices are dynamically selected as the training data points in each class, and the obtained scalable kernel convex hull can relieve class imbalance. Theoretical analysis and numerical experiments show that SKCHO-SVM has the distinctive ability of training time and classification performance. Keywords Online learning . Support vector machine . Scalable kernel convex hull . Network traffic classification
1 Introduction According to the 42nd National Internet development statistics report issued by China Internet Information Center, up to December 2018 the number of Internet users in China had reached 802 million [1]. With the increase of popularity of Internet, the scale of the Internet is becoming larger and larger; meanwhile, various new network applications and services are emerging. Network technology represented by wireless * Tongguang Ni [email protected] Xiaoqing Gu [email protected] Yiqing Fan [email protected] Weibo Wang [email protected] 1
School of Information Science and Engineering, Changzhou University, Changzhou 213164, China
2
Viterbi School of Engineering, University of Southern California, Los Angeles 90089, USA
3
School of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237, China
communications and networks has become one of the necessities in people’s daily life. Network traffic classification and application identification are the basis of solving many network management. They are of great significance to network security, intrusion detection and service quality guarantee. In the last two decades, several network traffic classification methods have been developed. The earliest traffic classifiers are mostly port-based classification methods that use the port number to identify applications [2]. The advantage of portbased classification methods is simple and fast. However, with more and more applications using dynamic port number, encryption
Data Loading...