摘要
随着社会网络的兴起,尤其是OLAP概念的提出,人们提出了很多对OLAP研究的算法,其中对图聚类算法的研究也引起了人们的广泛关注。但是这类算法大多数只是关注节点的属性或者节点之间的关系,而很少同时考虑到节点的属性和它们之间的关系。文中从这两个方面考虑的同时,在划分节点的时候更考虑到了要划分的节点在整个组中的紧密性,把网络中的模块化运用到节点的划分中,使得划分的结果更具有现实意义,而且很好地把Q函数的理论应用到社区的划分过程中,更加注重了单个节点对整个社区划分的影响,使得划分之后的各个子社区内部关系更加紧密。
With the rise of social networks,especially the concept of OLAP has been proposed,people make a lot of research on the OLAP algorithm,in which the graph clustering algorithm also attracts widespread attention.However,most of these algorithms are only concerned about attributes or relationships between nodes,rarely taking into account both the node's attributes and the relationships between them.In this paper,consider this problem,meanwhile even taking the problem of the tightness of the node which is in the division group into account,and also use the idea of the modular in the network on partitioning procedure,so that the results of partitioning will have a more realistic significance.And the Q-function theory is so well applied to the process of dividing the community that emphasize on the impact of individual nodes for the entire community,making each child after division within the community much closer.
出处
《计算机技术与发展》
2014年第6期99-102,106,共5页
Computer Technology and Development
基金
国家自然科学基金资助项目(61170052)