Abstract
In this paper, a new algorithm for incremental overlapped clustering, called Incremental Clustering by Strength Decision (ICSD), is introduced. ICSD obtains a set of dense and overlapped clusters using a new graph cover heuristic while reduces the amount of computation by maintaining incrementally the cluster structure. The experimental results show that our proposal outperforms other graph-based clustering algorithms considering quality measures and also show that ICSD achieves a better time performance than other incremental graph-based algorithms. © 2009 Springer-Verlag Berlin Heidelberg.
Cite
CITATION STYLE
Pérez Suárez, A., Martínez Trinidad, J. F., Carrasco Ochoa, J. A., & Medina Pagola, J. E. (2009). A new incremental algorithm for overlapped clustering. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5856 LNCS, pp. 497–504). https://doi.org/10.1007/978-3-642-10268-4_58
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
