大数据环境下网络异常数据准确检测仿真

发布时间：2018-04-14 10:36

本文选题：数据干扰 + 异常数据　；参考：《计算机仿真》2017年08期

【摘要】：大数据环境下对网络异常数据的准确检测能够确保网络系统的安全性。传统方法进行异常数据检测时,需要对全部数据的不同样本进行不同程度的扰动,过程较为复杂,且无法彻底抑制大规模冗余数据干扰,存在数据检测正确率低的问题。提出基于改进关联规则的大规模冗余数据干扰下的网络异常数据准确检测方法。定义异常数据的支持度和置信度,并给出最小支持度阈值和最小置信度阈值,利用混沌理论从网络全部数据序列中获取描述异常数据特征的混沌数据特征关联,定义正常数据间的紧密性和异常数据间的紧密度量,并得到当前各聚类内数据点的紧密性程度大小,利用数据特征聚类的紧密性与分离性方法对含有大规模冗余数据干扰的网络环境进行抑制,完成对网络异常数据准确检测。仿真结果表明,所提方法检测精确度高,可有效地保障网络安全稳定的运行。
[Abstract]:Big data environment to the network anomaly data accurate detection can ensure the security of the network system.When the traditional method is used to detect abnormal data, different samples of all data need to be disturbed to different degrees, the process is more complicated, and the interference of large-scale redundant data can not be completely suppressed, so there is a problem of low accuracy of data detection.An accurate detection method of network anomaly data based on improved association rules is proposed under the interference of large scale redundant data.The support degree and confidence degree of the abnormal data are defined, and the minimum support threshold and the minimum confidence threshold are given. The chaotic data feature association describing the abnormal data feature is obtained from all the data sequences of the network by using chaos theory.The compactness between normal data and abnormal data is defined, and the degree of tightness of data points in each cluster is obtained.By using the method of data clustering compactness and separability, the network environment with large scale redundant data interference is suppressed, and the network anomaly data is detected accurately.The simulation results show that the proposed method has high detection accuracy and can effectively ensure the safe and stable operation of the network.
【作者单位】：聊城大学东昌学院;
【分类号】：TP311.13;TP393.08

【相似文献】