面向网络状态的自适应用户行为评估方法

发布时间：2018-03-12 14:34

本文选题：用户行为评估　切入点：网络状态　出处：《通信学报》2013年07期 　论文类型：期刊论文

【摘要】：用户复杂的行为往往会导致网络状态出现波动,破坏网络平稳运行,由于指标及权重的主观性和静态性,传统评估方法难以准确衡量用户行为对网络状态变化的影响程度。因此引入粗糙集理论,构建一种面向网络状态的自适应用户行为评估方法,使用属性约简和属性重要度方法对用户行为和网络状态数据进行挖掘,分析用户行为与网络状态变化的关联程度,以此自适应构建评估指标及权重,并随用户行为变化而动态调整,从而准确地量化用户行为对网络状态变化的影响程度。实验结果表明,该评估方法有助于准确发现造成网络状态变化的用户及其行为,能够为加强对用户行为的管控提供有效支持。
[Abstract]:The complex behavior of the user often leads to the fluctuation of the network state and destroys the smooth operation of the network, because of the subjectivity and the static nature of the index and the weight. It is difficult for traditional evaluation methods to accurately measure the influence of user behavior on network state change, so a new adaptive user behavior evaluation method oriented to network state is constructed by introducing rough set theory. This paper uses attribute reduction and attribute importance method to mine user behavior and network state data, analyzes the correlation degree between user behavior and network state change, and adaptively constructs evaluation index and weight. The experimental results show that the proposed evaluation method is helpful to accurately detect the users and their behaviors that cause the network state change, and adjust dynamically with the change of user's behavior, thus accurately quantifying the influence of user's behavior on the network state change. Can provide effective support for strengthening the control of user behavior.
【作者单位】：东南大学计算机科学与工程学院;苏州科技学院电子与信息工程学院;
【基金】：国家重点基础研究发展计划(“973”计划)基金资助项目(2010CB328104) 国家自然科学基金资助项目(61070158,61003257,61070161,61070210) 国家高技术研究发展计划(“863”计划)基金资助项目(2013AA013503) 高等学校博士点学科专项科研基金资助项目(20110092130002) 江苏省网络与信息安全重点实验室基金资助项目(BM2003201) 教育部计算机网络与信息集成重点实验室基金资助项目(93K-9)~~
【分类号】：TP393.0

【参考文献】