信息工程大学学报2025，Vol.26Issue(5)：575-584,10.DOI:10.3969/j.issn.1671-0673.2025.05.011
基于kNN检索的网络流量检测域适应增强方法
Domain Adaptation Enhancement Method for Network Traffic Detection Based on kNN Retrieval
摘要
Abstract
To address the issue that the insufficient generalization ability of the model due to the data distribution difference between the target domain and the source domain in network traffic detection leads to declined detection performance,a non-parametric traffic detection method with enhanced do-main adaptation ability is proposed.Specifically,during the model fine-tuning stage,a dictionary is constructed by integrating the deep representations of training samples with their labels.In the detec-tion phase,the dictionary is queried to correct the prediction results of the parameterized detection model based on the probability distribution of the k most similar samples to the sample under test.This approach achieves stable and robust detection results without increasing training sample size or model training overhead and possesses high interpretability.Extensive experiments on cross-domain sce-narios with imbalanced distributions(e.g.,USTC-TFC2016 dataset)demonstrate that the model's adap-tation ability and detection performance are significantly enhanced by the method,and the effective-ness and robustness of the method are further verified under different new domain data distributions.关键词
网络流量检测/非参数化方法/域适应/可解释性/跨域检测/不平衡数据分布Key words
network traffic detection/non-parametric method/domain adaptation/explainability/cross-domain detection/imbalanced data distribution分类
计算机与自动化引用本文复制引用
孙剑文,张斌,张昊..基于kNN检索的网络流量检测域适应增强方法[J].信息工程大学学报,2025,26(5):575-584,10.基金项目
信息工程大学密码工程学院研究生创新基金资助项目(2019f113) （2019f113）
