sandesh20lamichhane/when-should-ids-adapt
- 类型:github
- 标识:sandesh20lamichhane/when-should-ids-adapt
- 链接:https://github.com/sandesh20lamichhane/when-should-ids-adapt
- 主题:engineering
- 主分类:evaluation
- 形态:framework
- 分类:academic-writing
- Stars:0
- 周增:+0
- 语言:Jupyter Notebook
- 最近提交:2026-08-19
- 简介:A reproducible research framework for investigating when an Intrusion Detection System (IDS) should adapt under temporal distribution shift. The project evaluates a staged decision pipeline combining drift screening, novelty detection, and cost-aware adaptation triggers to determine whether retraining is necessary and beneficial.
- 上次采集:2026-08-19
- 首次采集:2026-08-19
- 学术状态:accepted
- 学术阶段:research
- 学术主任务:reproducibility
- 学术方向:misc
- 学术用途:research-design
- 学术相关度:80
- 学术判定:deterministic
- 学术证据:anchor:research; query:reproducible-research; task:reproducibility:reproducible research
- 学术复核时间:2026-08-19T13:30:02+08:00
- 待LLM分类:否
- 成熟度:experimental
- 简介中文:一个用于研究入侵检测系统 (IDS) 在时间分布漂移下何时应进行适应的可复现研究框架。该项目评估了一个分阶段决策流水线,结合漂移筛查、新颖性检测与成本感知的适应触发器,以判断重训练是否必要且有益。
- 来源文件:
- [GitHub Search]