Now 6454 visitors
Today:0 Yesterday:0
Total: 131455 216S 51P 5R
2026-10-21, Week 43
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


















Online Q&A Session
Paper Number Author:
Paper Title
Keyword
Q&A Number **Select your question number !!
Questioner eMail:
Question
Answer
by
Author
Save Q&A

* Edit or answer any Q&A by selecting Q&A number Hyper Link below + Write button (Save)
ICACT20210143 Question.1
Questioner: xiaochunyin@wfust.edu.cn    2021-02-09 오후 4:06:05 		ICACT20210143 Answer.1
Answer by Auhor st.lzg@163.com   2021-02-09 오후 4:06:05
Chrome Click!!
Could authors give more description about "Enhanced information entropy"? Thanks for questions. In order to use entropy in feature extraction, conditional probability is introduced into entropy definition. Now entropy is calculated based on event probability. Every feature value is treated as event. Thus, event entropy means the distribution of value. And we define S(i, l), which means the entropy of i-th feature value in single classification of l.
ICACT20210143 Question.2
Questioner: xiaochunyin@wfust.edu.cn    2021-02-09 오후 4:11:23 		ICACT20210143 Answer.2
Answer by Auhor st.lzg@163.com   2021-02-09 오후 4:11:23
Chrome Click!!
Did authors have a experiment in simulated IoT lab? If yes, what is the result? Yes. we simulated in IoT lab. During experiment, we hacked randomly 1k endpoints/users as bots, and start to attack our cloud data center with the peak rate of 50Kb/sec and the attacking period of 500ms. We collected all raw TCP dump data with same dimensions and with extra source and destination IP. The comparations were almost the same with the experiment result of "KDD CUP-99 10% dataset".
ICACT20210143 Question.3
Questioner: xiaochunyin@wfust.edu.cn    2021-02-09 오후 4:36:52 		ICACT20210143 Answer.3
Answer by Auhor st.lzg@163.com   2021-02-09 오후 4:36:52
Chrome Click!!
Could authors light us how to plan to generate adversarial samples in near future? As you known, the LDDoS is a time-related attack. We plan to learn its time-domain feature by LSTM. And then feed them into GAN network. We expect that we have a new and good finding.