环形网络大数据关联特征无规则挖掘算法仿真

Simulation of Random Mining Algorithm for Big Data Association Features in Ring Networks

环形网络的大数据特征相似度较高,挖掘任务量大,数据关联规则挖掘的阈值以及近邻个数具有随机性,加大了大数据特征挖掘的难度。为此提出环形网络大数据关联特征无规则挖掘算法。构建语义概念树,利用该模型分析环形网络分布结构特征。采用信息融合方法匹配环形网络大数据的分块结构。基于匹配结果对重构后的环形网络完成关联规则特征提取,以提取的关联特征用作特征信息素,实现环形网络大数据关联特征的无规则挖掘。仿真中测试上述方法的挖掘准确率测试、挖掘用时指标,实验结果表明了上述方法的大数据关联特征无规则挖掘效率高,可靠性强。

The similarity among big data features in ring networks is too high,the threshold and the number of nearest neighbors mined by data association rule are random,which increases the difficulty of big data feature mining.Therefore,an irregular mining algorithm for big data association features in the ring network was proposed.Firstly,a semantic concept tree was constructed.Based on this model,the structural characteristics of ring network distribution were analyzed.Then,the blocks of big data in ring networks were matched by the information fusion method.Based on the matching results,the association rule features of the reconstructed ring network were extracted as feature pheromones,thus realizing the irregular mining for big data association features of the ring network.In the simulation,the mining accuracy and mining time were tested.Test results show that the proposed method has high efficiency and strong reliability.