摘要
传统方法在监测医院网络流量时存在监测精度低、监测时间长和监测漏报率高等问题,提出基于决策树模型的医院网络流量监测方法。采集医院网络流量数据;采用多尺度分析方法提取网络流量特征,增强网络流量在分类过程中抗干扰能力;利用决策树模型对医院网络流量进行分类,并引入滑动窗口技术对分类效果进行提升,通过改进VFDT算法在决策树节点中生成新的替代子树,同时维护决策树的连续性,从中建立二叉排序属性树,使新样本在插入时只需更新一个节点,减少决策树分类网络流量时间,完成医院网络流量监测。通过对医院网络流量监测精度、监测时间和监测漏报率的测试,验证该方法有效性。
When traditional methods are used to monitor network traffic,there are some problems such as low accuracy,long monitoring time and high rate of missing reports.A hospital network traffic monitoring method based on decision tree model is proposed.It collects the hospital network traffic data,and obtains the characteristics according to the collected data.The multi-scale analysis method is used to extract the characteristics of network traffic to enhance the anti-interference ability in the process of classification;The hospital network traffic is classified by the decision tree model,and the sliding window technology is introduced to improve the classification effect.By improving the VFDT algorithm,a new alternative subtree is generated in the decision tree node.At the same time,the continuous attributes of the decision tree are maintained,from which a binary sorting attribute tree is established,so that only one node needs to be updated when new samples are inserted.It reduces the network traffic time of decision tree classification,and completes hospital network traffic monitoring.In experiments,the effectiveness of this method is verified by testing the monitoring accuracy,monitoring time and monitoring missing rate of hospital network traffic.
作者
钟伟华
黄达辉
王欣霖
于欣宁
ZHONG Weihua;HUANG Dahui;WANG Xinlin;YU Xinning(Meizhou Acadency of Medical Sciences,Meizhou People’s Hospital(Huang Tang Hospital),Meizhou 514031,China)
出处
《微型电脑应用》
2023年第4期91-93,97,共4页
Microcomputer Applications
关键词
决策树模型
医院网络流量监测
滑动窗口技术
二叉排序属性树
decision tree model
hospital network traffic monitoring
sliding window technology
binary sorting attribute tree
