Feature Selection Method for the Classification of Traffic in SDN 


Vol. 44,  No. 1, pp. 106-116, Jan.  2019
10.7840/kics.2019.44.1.106


PDF
  Abstract

Nowadays, various Internet of Things (IoT) devices are widely used, generating tremendous amount of traffic. They not only slow the transmission speed of the network, but also make it difficult to guarantee high QoS. The SDN technology had been introduced as a solution to these problems. SDN is used in a large-scale network environment because it can efficiently manage the network by separating the control plane and data plane. This paper proposes a novel feature selection algorithm to efficiently classify various internet traffics in SDN environment. The shortcoming of the existing filter-based feature selection approach is low classification accuracy if the number of features is small. In order to solve such problem, a novel feature selection scheme is proposed which employs the weight-based chi2-square test algorithm. The experimental results with three datasets reveal that the proposed scheme outperforms two popular existing feature selection algorithms in terms of classification accuracy and F1 score.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

H. Lim, K. Kim, B. Lee, H. Youn, "Feature Selection Method for the Classification of Traffic in SDN," The Journal of Korean Institute of Communications and Information Sciences, vol. 44, no. 1, pp. 106-116, 2019. DOI: 10.7840/kics.2019.44.1.106.

[ACM Style]

Hwan-hee Lim, Kyung-tae Kim, Byung-jun Lee, and Hee-yong Youn. 2019. Feature Selection Method for the Classification of Traffic in SDN. The Journal of Korean Institute of Communications and Information Sciences, 44, 1, (2019), 106-116. DOI: 10.7840/kics.2019.44.1.106.

[KICS Style]

Hwan-hee Lim, Kyung-tae Kim, Byung-jun Lee, Hee-yong Youn, "Feature Selection Method for the Classification of Traffic in SDN," The Journal of Korean Institute of Communications and Information Sciences, vol. 44, no. 1, pp. 106-116, 1. 2019. (https://doi.org/10.7840/kics.2019.44.1.106)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880