摘要
针对传统的决策树分类特征滤波方法识别性能不好的问题,提出一种基于决策树局部时间尺度特征提取的大型网络数据库中的危险数据识别方法,构建大型网络数据库的危险数据通信传输模型,设计危险数据决策树模型,通过建模,采用信号处理方法实现局部时间尺度特征提取,以此为依据实现对危险数据的准确识别。仿真结果表明,采用该算法能有效地实现危险数据的识别,准确识别概率提高显著,保证了大型网络数据库的安全运行。
In view of low identification performance of traditional decision tree classification feature filter method, this paper puts forward an identification method of dangerous data in large-scale network database extracted based on decision tree local time scale ~ature. The trartsmission model of dangerous data in large-scale network database is set up, and the decision tree model of dangerous data is designed. By modeling, the signal processing method is used to perform local time scale feature extraction for exact identification of dangerous data. The simulation results show that this method can effectively implement dangerous data identification with higher identification probability and ensure safe operation of large-scale network database.
出处
《计算机与网络》
2016年第3期105-107,共3页
Computer & Network
关键词
网络数据库
特征提取
危险数据
识别
network database
feature extraction
risk data
identification
