A new blockmodeling based hierarchical clustering algorithm for web social networks

摘要: Cluster analysis for web social networks becomes an important and challenging problem because of the rapid development Internet community like YouTube, Facebook TravelBlog. To accurately partition networks, we propose a hierarchical clustering algorithm called HCUBE based on blockmodeling which is particularly suitable with complex link relations. uses structural equivalence to compute similarity among pages reduces large incoherent network into set smaller comprehensible subnetworks. actually bottom-up agglomerative inter-connectivity closeness clusters group structurally equivalent in effective fashion. In addition, address preliminaries proposed theoretical foundations algorithm. order improve efficiency HCUBE, optimize it by reducing its time complexity from O(|V|^2) O(|V|^2/p), where p constant representing number initial partitions. Finally, conduct experiments real data results show that at partitioning compared Chameleon k-means algorithms.