计算机系统应用 ISSN 1003-3254, CODEN CSAOBN E-mail: ******@
Computer ication contributions for each category and remove irrelevant feature
attributes. Then, the AMB algorithm is adopted to delete redundant features in the subset of candidate features. Finally,
the Wrapper approach based on the classification algorithm is employed to determine the final feature preference.
The experimental results show that the accuracy of the features selected under this method for type identification of the
network device operating system has been improved compared with classical feature selection methods, and the recall rate
on small class data has also been raised.
Key words: feature selection; network traffic; symmetric uncertainty (SU); approximate Markov blanket (AMB); network
devices identification; machine learning
①基金项目:兴辽英才计划(XLYC2019019)
收稿时间:2021-06-07;修改时间:2021-07-07, 2021-07-20;采用时间:2021-07-27; csa 在线出版时间:2022-03-22
Software Technique*Algorithm 软件技术•算法 281计算机系统应用 -s- 2022年第31卷第4期
随着联网设备的增多,越来越多的网络设备暴露 提出了一种Filter-Wrapper混合的特征选择模型,其先
在公网内,网络空间扫描工具Shodan甚至可以轻而易 利用信息增益对有分类贡献的特征进行性能评估排序,
举获取到隐私家用摄像头
基于SU和AMB的网络流量特征选择算法 来自淘豆网www.taodocs.com转载请标明出处.