Selection of the number of clusters is a crucial problem in clustering. Conventionally, it was effected via cost function based criteria such as AIC and MDL. In this paper we empirically investigate automatic selection of the number of clusters via BYY harmony empirical learning. Results of experiments show that the true number of clusters can be automatically obtained during BYY harmony empirical learning. It is superior to conventional methods in that it needs much less computational cost. © Springer-Verlag 2004.
CITATION STYLE
Hu, X., & Xu, L. (2004). Automatic cluster number determination via BYY harmony learning. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3173, 828–833. https://doi.org/10.1007/978-3-540-28647-9_136
Mendeley helps you to discover research relevant for your work.