Encyclopedia of Machine Learning

2010 Edition
| Editors: Claude Sammut, Geoffrey I. Webb

Cluster Editing

Reference work entry
DOI: https://doi.org/10.1007/978-0-387-30164-8_121

The Cluster Editing problem is almost equivalent to Correlation Clustering on complete instances. The idea is to obtain a graph that consists only of cliques. Although Cluster Deletion requires us to delete the smallest number of edges to obtain such a graph, in Cluster Editing we are permitted to add as well as remove edges. The final variant is Cluster Completion in which edges can only be added: each of these problems can be restricted to building a specified number of cliques.

Copyright information

© Springer Science+Business Media, LLC 2011