|
Abstract:
|
Most generative models for clustering implicitly assume that the number of data points in each cluster grows linearly with the total number of data points. Finite mixture models, Dirichlet process mixture models, and Pitman-Yor process mixture models make this assumption, as do all other infinitely exchangeable clustering models. However, this assumption is undesirable in applications for certain tasks such as record linkage and community detection. These tasks therefore require models that yield clusters whose sizes grow sublinearly with the size of the data set. We address this requirement by defining the microclustering property. We introduce one model satisfying this property, showing its promise to both official statistics and simulated data.
|
