基于小样本学习的网络异常流量检测
Abnormal Network Traffic Detection Based on Small Sample Learning
李荣宽 1丁乙 1王寒凝 2贺宁3
作者信息
- 1. 电科云(北京)科技有限公司 北京 100041
- 2. 解放军61932部队 北京 100000
- 3. 东南大学网络空间安全学院 南京 211189
- 折叠
摘要
网络结构具有较高复杂性,因此导致各种异常流量现象层出不穷,其中包括一些标注样本极少的新型异常流量类型.为了有效识别标注样本量极少的异常情况,设计了一种基于小样本学习的网络异常流量检测方法.该方法利用基于小样本的迁移学习技术识别异常流量,从而确保了网络安全.
Abstract
The high complexity of network structure leads to various abnormal traffic phenomena,in-cluding some new abnormal traffic types with few labeled samples.In order to effectively identify the abnormal situations with few labeled samples,an abnormal network traffic detection method based on small sample learning is designed.The method uses the transfer learning technology based on small samples to identify the abnormal traffic.Thus,it can ensure the network security.
关键词
小样本/迁移学习/网络异常流量Key words
small sample/transfer learning/abnormal network traffic引用本文复制引用
出版年
2024
NETL
NSTL
万方数据