Chinese Whispers is an unsupervised clustering algorithm that is able to cluster a graph into subgraphs. The number of subgraphs is determined automatically.
Chinese Whispers can be used in many applications where the number of cluster is not determined a priori and varies between the nodes. In the JoBimText framework, it is used to cluster similar terms into sense clusters.
You can view the javadoc documentation here: Chinese Whispers doc
The algorithm is described in:
Biemann, C. (2006): Chinese Whispers – an Efficient Graph Clustering Algorithm and its Application to Natural Language Processing Problems. Proceedings of the HLT-NAACL-06 Workshop on Textgraphs-06, New York, USA (pdf ) (ppt )