加权社会网络低维冗余数据快速挖掘算法仿真

Simulation of Fast Mining Algorithms for Low-Dimensional Redundant Data in Weighted Social Networks

在进行加权社会网络低维冗余数据挖掘时,由于加权社会网络稳定性低,且挖掘聚类效果差,导致挖掘耗时长与挖掘精度较低,因此设计加权社会网络低维冗余数据快速挖掘算法。构建加权社会网络模型,提升该网络稳定性的同时对收集到的数据进行可视化分析。通过特征选择获取数据冗余特征,计算出低维冗余聚类数据的支持度,利用支持度与可信度对低维冗余数据关联规则进行评价,并按照直接属性对其限制,大幅度减少无用规则的产生。通过属性位复用方法建立候选区域,生成关联规则集,对符合关联规则集的低维冗余数据进行聚类,从而实现数据的快速挖掘。仿真结果表明:所提方法的挖掘聚类效果好、挖掘精度高、耗时短,具有可行性。

When we mine low-dimensional redundant data in a weighted social network,the mining accuracy is low due to the low stability of weighted social network and poor mining clustering effect.Therefore,a fast mining algorithm for low-dimensional redundant data in a weighted social network was designed.The weighted social network model was constructed to improve the stability of the network.Meanwhile,the data was analyzed visually.Then,redundant features were obtained through feature selection,and the support degree of low-dimensional redundant clustering data was calculated.Moreover,the association rules of low-dimensional redundant data were evaluated by the support degree and reliability,and they were limited according to direct attributes,so as to greatly reduce the useless rules.Furthermore,the attribute bit multiplexing method was adopted to build candidate regions,and thus to generate the association rule sets.Finally,low-dimensional redundant data conforming to association rule sets were clustered.Thus,we realized fast data mining.Simulation results show that the proposed method has a good mining clustering effect,high mining accuracy,short running time,and good feasibility.