摘要
为了解决复杂信息网络中多链接高维数据聚类难以处理且效率较低问题,提出了一种新颖的基于高阶张量分析方法和模块化网络分析方法相结合的链接聚类算法(modularity-clustering-HOOI,MCHOOI)。利用模块化方法分析网络,利用张量的形式表示多维的复杂的多链接数据,利用Tucker张量分解的方法进行降维处理,有效地降低了算法的时间和空间复杂度,并在实际网络环境下,通过实验验证了算法的有效性和健壮性。
Because the multi-link high-dimensional data clustering problem of complex information networks was difficult to handle,this paper proposed a novel link clustering algorithm MCHOOI based on higher order tensor analysis methods and modularity network analysis.Used modularity approach to analysis networks,used multi-dimensional tensor expressed in the form of complex multi-link data,used Tucker tensor decomposition method to reduce the dimensions of the data,the time and space complexity of the algorithm.The effectiveness and robustness of the algorithmic was tested in complex network environment.
出处
《计算机应用研究》
CSCD
北大核心
2011年第3期833-837,共5页
Application Research of Computers
基金
国家自然科学基金资助项目(60675030
60875029)