Generalized Negative Binomial Processes and the Representation of Cluster Structures
arXiv:1310.1800
Abstract
The paper introduces the concept of a cluster structure to define a joint distribution of the sample size and its exchangeable random partitions. The cluster structure allows the probability distribution of the random partitions of a subset of the sample to be dependent on the sample size, a feature not presented in a partition structure. A generalized negative binomial process count-mixture model is proposed to generate a cluster structure, where in the prior the number of clusters is finite and Poisson distributed and the cluster sizes follow a truncated negative binomial distribution. The number and sizes of clusters can be controlled to exhibit distinct asymptotic behaviors. Unique model properties are illustrated with example clustering results using a generalized Polya urn sampling scheme. The paper provides new methods to generate exchangeable random partitions and to control both the cluster-number and cluster-size distributions.
arXiv comment30 pages, 8 figures
References in corpus (6)
- Model Selection in Linear Mixed Models
- Beta-Negative Binomial Process and Poisson Factor Analysis
- MCMC for Normalized Random Measure Mixture Models
- Stochastic expansions using continuous dictionaries: Lévy adaptive regression kernels
- Combinatorial clustering and the beta negative binomial process
- Characterizations of exchangeable partitions and random discrete distributions by deletion properties