基于相空间融合的异构网络数据拓扑重构技术

Dynamic Topology Reconstruction Technology of Heterogeneous Network Data Based on Phase Space Fusion

为了提高分布式异构网络数据动态拓扑重构能力,提出基于相空间融合的异构网络数据动态拓扑重构算法。构建模糊层次空间异构网络数据库的异构存储结构模型,采用网格分块特征检测和特征重组技术实现对模糊层次空间异构网络数据库的动态拓扑设计,挖掘模糊层次空间异构网络数据库的关联规则特征分布集,采用相空间融合和模糊C均值聚类方法,实现对模糊层次空间异构网络数据库的统计信息融合聚类。根据数据的动态拓扑重构聚类结果,结合随机自适应调度和子空间压缩方法,实现对模糊层次空间异构网络数据库的动态拓扑重构,提高模糊层次空间异构网络数据库查询数据的特征聚类性。采用Matlab语言实现仿真分析,仿真结果表明,采用该方法进行异构网络数据动态拓扑重构的查准率较高,查全性较好,异构网络数据动态拓扑输出的聚类性较强。

In order to improve the ability of dynamic topology reconstruction of distributed heterogeneous network data,an algorithm of dynamic topology reconstruction of heterogeneous network data based on phase space fusion is proposed.The heterogeneous storage structure model of heterogeneous network database in fuzzy hierarchical space is constructed.The dynamic topology design of heterogeneous network database in fuzzy hierarchical space is realized by using grid block feature detection and feature recombination technology.The association rule feature distribution set of heterogeneous network database in fuzzy hierarchical space is mined.The statistical information fusion clustering of heterogeneous network database in fuzzy hierarchical space is realized by using phase space fusion and fuzzy C-means clustering methods.According to the clustering results of data dynamic topology reconstruction,combining random adaptive scheduling and subspace compression methods,the dynamic topology reconstruction of heterogeneous network database in fuzzy hierarchical space is realized,and the feature clustering of query data in heterogeneous network database in fuzzy hierarchical space is improved.Matlab language is used to realize simulation analysis,and the simulation results show that this method has higher precision,better precision rate and stronger clustering of heterogeneous network data dynamic topology output.