DPGMM.js is a JavaScript implementation of a Dirichlet process gaussian mixture model (DPGMM). Live Demo
The Dirichlet process is a stochastic process used in Bayesian nonparameteric models of data, particularly in infinite mixture models. It can be used to build clustering algorithms which do not make a-priori assumptions about the number of components in the data.
The project relies on the following dependencies:
- Neal, Radford M. "Markov chain sampling methods for Dirichlet process mixture models." Journal of computational and graphical statistics 9.2 (2000): 249-265.
- Conjugate Bayesian analysis of the Gaussian distribution
- Infinite Mixture Models with Nonparametric Bayes and the Dirichlet Process
- Collapsed Gibbs Sampling for Dirichlet Process Gaussian Mixture Models
- Overview of Cluster Analysis and Dirichlet Process Mixture Models