Co-clustering documents and words using bipartite spectral graph partitioning
Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining
Biclustering Algorithms for Biological Data Analysis: A Survey
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
| Hi-index | 0.00 |
Discovering communities from networks is one of the important techniques for intelligent Web interaction. Most of the previous methods for discovering communities are for homogeneous networks composed of only one type of vertices. In real world situations, however, there are many heterogeneous networks composed of more than one types of vertices. This paper describes our attempts for discovering, visualizing and evaluating communities from bipartite networks. A biparite network is projected to two homogeneous networks, and communities are discovered from each of the networks. The communites are visualized on two windows in order to clarify the correspondence between communities of different vertex types. Discovered communities are then evaluated by bipartite modularity. These attempts will clarify the overall structure of given networks and contribute to the interactive exploration of online activities.