Now 6931 visitors
Today:0 Yesterday:0
Total: 153698 212S 50P 4R
2026-10-20, 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




















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)
ICACT20260037 Question.1
Questioner: ygt991129@mail.ustc.edu.cn    2026-02-09 오후 6:21:55
ICACT20260037 Answer.1
Answer by Auhor jianms@nfu.edu.tw   2026-02-09 오후 6:21:55
Chrome Click!!
How can the voting mechanism be optimized to effectively handle hybrid network attacks, reduce classification confusion, and improve detection accuracy? By voting mechanism, we try to integrate the multiple algorithms for a decision. If the estimation done by different methods results in the various accuracy. We focus on the potential risk, not on the known security risk. In other words, we dont really check the accuracy
ICACT20260037 Question.2
Questioner: ygt991129@mail.ustc.edu.cn    2026-02-09 오후 6:24:15
ICACT20260037 Answer.2
Answer by Auhor jianms@nfu.edu.tw   2026-02-09 오후 6:24:15
Chrome Click!!
In resource-constrained edge environments, how can the deployment speed and resource efficiency of containerized AI modules be further optimized to support real-time threat response? There are two different issues: one is for AI modile deployment, another is bait system deployment. Since we need to analyse the data packets, it is difficult to be a real time system. For bait system, the latency of the network xould be the time for bait system deployment.
ICACT20260037 Question.3
Questioner: ygt991129@mail.ustc.edu.cn    2026-02-09 오후 6:25:36
ICACT20260037 Answer.3
Answer by Auhor jianms@nfu.edu.tw   2026-02-09 오후 6:25:36
Chrome Click!!
How can AI modules be enhanced to better recognize unknown or novel network attacks through incremental learning, thereby improving the systems adaptive capability? Instead, we only cluster the instructions into normal or abnormal. Hence, any connection without normal instructions is abnormal.
ICACT20260037 Question.4
Questioner: zhaoyb@dsp.ac.cn    2026-02-10 오후 4:37:14
ICACT20260037 Answer.4
Answer by Auhor jianms@nfu.edu.tw   2026-02-10 오후 4:37:14
Chrome Click!!
Besides phishing, ransomware, and SQL injection, are there other emerging cyber-attack types that your research specifically targets? Why did you prioritize these three attack categories in your initial analysis? In this research, we will take these instructions as the simple examples. Indeed, we consider the normal instructions more than you abnormal instructions. The normal instructions are defined as the known instructions without security risk. Hence, by excluding the normal instructions and operations, we could reduce the risk.
ICACT20260037 Question.5
Questioner: zhaozx@dsp.ac.cn    2026-02-24 오전 7:30:28
ICACT20260037 Answer.5
Answer by Auhor jianms@nfu.edu.tw   2026-02-24 오전 7:30:28
Chrome Click!!
You mentioned that traditional systems rely on static signatures and struggle with zero-day threats. Could you specify what types of obfuscation techniques (e.g., code encryption, traffic mixing patterns) are most challenging for static signature-based systems, and how they specifically evade detection? Sorry, instead of the known security issue, we would rather compare the intruction or operation "NOT" as the normal operation corresponding to the current service. Any operation NOT as the normal operation would be defined as the abnormal connection.