Title :
Regularized modularity eigenmap for community discovery with side information
Author :
Yoshida, Tetsuya
Author_Institution :
Grad. Sch. of Inf. Sci. & Technol., Hokkaido Univ., Sapporo, Japan
Abstract :
We propose a method for community discovery with side information based on a regularized modularity eigenmap. Community discovery has been conducted mostly based solely on the connectivity relation among nodes in a social network. However, the connectivity may change in time, or some links might be missing. Even when the connectivity relation in a network is only partially available, if other information about the network is available, it can be exploited as auxiliary or side information for community discovery. Our approach constructs a graph structure based on the side information so that both connectivity relation and side information can be uniformly dealt with in terms of graph representation. An objective function with a regularization term for the side information is proposed based the modularity matrix of a network. Extensive experiments are conducted over social network datasets and comparison with several state-of-the-art methods is reported.
Keywords :
eigenvalues and eigenfunctions; graph theory; matrix algebra; social networking (online); community discovery; connectivity relation; graph representation; graph structure; modularity matrix; regularization term; regularized modularity eigenmap; side information; social network datasets; Abstracts; Argon; Communities; Machine learning; Social network services; Symmetric matrices; Vectors; community discovery; eigenmap; embedding; side information;
Conference_Titel :
Granular Computing (GrC), 2011 IEEE International Conference on
Conference_Location :
Kaohsiung
Print_ISBN :
978-1-4577-0372-0
DOI :
10.1109/GRC.2011.6122697