基于不满意度的网络安全模型

发布时间：2018-09-11 06:51

【摘要】：提出了一种基于不满意度的网络安全模型,主要功能是帮助用户在网络环境中正确地选择交易对象,屏蔽恶意节点,基于不满意度(degree of dissatisfaction,DoD)对交易节点进行分类控制。节点的不满意度定义为该节点属于恶意节点集的概率。a)使用粗糙集(rough set)模块与Bayesian学习器计算节点的不满意度,依据节点的交易历史记录计算节点的本地不满意度(local DoD,LDoD),依据反馈推荐意见计算推荐不满意度(recom-mended DoD,RDoD),基于不满意度将节点划分为可信任节点、陌生节点、恶意节点等不同的类型;b)基于推荐意见的信息熵(information entropy)计算其可信度,对反馈推荐意见进行综合。实验表明,与已有的安全模型相比,提出的安全管理模型对恶意节点具有更高的检测率,具有更满意的交易成功率。
[Abstract]:In this paper, a network security model based on dissatisfactory degree is proposed. The main function of this model is to help the user select the transaction object correctly in the network environment, shield the malicious node, and classify the transaction node based on the dissatisfactory (degree of dissatisfaction,DoD. The dissatisfaction of a node is defined as the probability that the node belongs to a malicious set of nodes. A) using the rough set (rough set) module and the Bayesian learner to calculate the node's dissatisfaction, Based on the transaction history of the node, the local dissatisfaction (local DoD,LDoD) of the node is calculated, and the recommended dissatisfaction (recom-mended DoD,RDoD) is calculated according to the feedback recommendation. Based on the dissatisfaction, the node is divided into trusted nodes and unfamiliar nodes. Different types of malicious nodes, such as malicious nodes, calculate their credibility based on the information entropy (information entropy) of the recommendation, and synthesize the feedback recommendations. Experimental results show that the proposed security management model has a higher detection rate and a more satisfactory transaction success rate than the existing security model.
【作者单位】：西安理工大学计算机学院;西安交通大学电子信息工程学院;
【基金】：国家自然科学基金资助项目(60873071)
【分类号】：TP393.08

【参考文献】