多层物联网下的非法信息快速定位方法研究

Study on Illegal Information Fast Positioning Method Under Multilayer Interment of Things

在对多层物联网下非法信息进行快速定位的过程中,容易出现实际问题中用户的行为模式存在多样性与不可预知性的情况,导致传统方法由于采用提前设定模式进行学习,无法有效实现非法信息的快速定位,提出一种基于K-均值聚类的多层物联网下非法信息快速定位方法,分析了多层物联网的结构,随机选择若干数据点作为初始簇中心,将集合中所有数据点分配至和其最近的簇中心所在的类簇中,形成初始分布。分配完成后对各簇中心进行更新,不断进行数据划分,直至簇中心不再发生改变,聚类准则函数收敛。通过评价指标获取合理的聚类结果,从而完成多层物联网下非法信息的快速定位。仿真实验表明,所提方法具有很高的准确性及高效性。

Under the multilayered Iot illegal information for rapid positioning in the process, it's easy to have a real problem existed in the user's behavior patterns of diversity and unpredictability, lead to the traditional methods with early set model for learning, unable to effectively achieve rapid positioning of illegal information, proposes a k-means clustering are based on multilayer iot of illegal information fast positioning method, analyzes the multi-layer structure of the Internet of things, random number of data points as the initial cluster centers, the collection of all data points assigned to and its nearest cluster center of the cluster, the formation of the initial distribution.Assigned to update of each cluster center, after the completion of ongoing data partitioning, until the cluster center no longer change, clustering convergence criterion function.Through the evaluation index to obtain reasonable clustering results, so as to complete fast localization of multilayer illegal information under the Internet of things.The simulation results show that the proposed method has high accuracy and high efficiency.