www.icact.org

Member Login

TACT Journal Homepage

Call for Paper

Committee

Find My Paper

Author Homepage

Statistics & History

Welcome Message

Paper Procedure

FAQ

Paper Archives

Outstanding Papers

Program & Proceedings

Presentation Platform

Hotel & Travel Info

Photo Gallery

Scheduler Login

Seminar

Archives Login

Sponsors

Virual Oral Question and Answer


IEEE/ICACT20230225 Question.2 Questioner: haodp@dau.edu.vn 2023-02-20 오후 8:18:43	IEEE/ICACT20230225 Answer.2 Answer by Auhor 6025211020@mhs.its.ac.id 2023-02-20 오후 8:18:43
What are the most effective feature reduction techniques for botnet activity detection in different types of network traffic, such as network flow data and system log data?	From the experimental results, reducing the feature dimensions into one feature using network flow data can conclude that Fast ICA proved to be the most effective for classifying botnet traffic and normal traffic. But other dimensional reduction techniques are still effective. Because the effectiveness of dimensional reduction techniques depends on many factors, including the relevance of the extracted features, the number of features included, and reduced to how many features. So analyzing these factors is needed to increase the effectiveness and efficiency of the dimensional reduction technique. Unfortunately, the experiment only uses network flow data, so it is not easy to provide an analysis related to the log data system because these two data have different characteristics. Analyzing the impact of the dimensional reduction technique on the log data system is also a good idea to do in the future. Thanks for the inspiring question.

Select Voice