大数据驱动下网络入侵信号提取检测仿真

发布时间：2018-02-25 00:10

本文关键词： 大数据驱动网络入侵提取检测隐马尔可夫模型　出处：《计算机仿真》2017年09期 　论文类型：期刊论文

【摘要】：大数据驱动下对网络入侵信号的提取检测,能够充分的保障大数据驱动下的网络安全。对网络入侵信号的提取检测,需要获取网络入侵提取的属性核,对网络入侵提取结果数据进行分类,完成入侵检测。传统方法定义网络入侵提取结果选取状态的分类的熵,给出各个熵的信息增益,但忽略了对网络入侵提取结果数据的分类,导致提取检测精度偏低。提出基于粗糙集-决策树结合的大数据驱动下的网络入侵信号提取检测模型。模型先利用粗糙集对大数据驱动下的网络中各个提取数据集中属性对应的取值进行离散化,获取网络入侵提取的属性核,利用决策树对新的网络入侵提取结果数据进行分类,初始化HMM模型的参数,将网络入侵信号提取检测的特征向量输入HMM,组建大数据驱动下的网络入侵信号提取检测模型。实验结果表明,所提模型建模精度较高,为保障大数据驱动下的网络安全奠定了基础。
[Abstract]:The extraction and detection of the network intrusion signal driven by big data can fully guarantee the network security driven by big data. The extraction and detection of the network intrusion signal need to obtain the attribute core of the network intrusion extraction. The traditional method defines the entropy of the classification of the selected state of the network intrusion extraction result, and gives the information gain of each entropy, but neglects the classification of the network intrusion extraction result data. The model of extracting and detecting network intrusion signal driven by big data based on rough set and decision tree is put forward. Firstly, rough set is used to extract every data in the network driven by big data. The values corresponding to the attributes in the set are discretized, The attribute kernel of network intrusion extraction is obtained, the new network intrusion extraction result data is classified by decision tree, and the parameters of HMM model are initialized. The feature vector of network intrusion signal extraction and detection is input into HMM.The model of network intrusion signal extraction and detection driven by big data is constructed. The experimental results show that the proposed model has high modeling accuracy. It lays a foundation for guaranteeing the network security driven by big data.
【作者单位】：内蒙古财经大学计算机信息管理学院;
【基金】：内蒙古自治区教育科学规划课题《高等教育发展性学生评价的研究》内教科规办强字[2011]01
【分类号】：TP393.08

【相似文献】