Chinese restaurant franchise
Summary
The Chinese Restaurant Franchise (CRF) is the predictive process for a hierarchical partitioning (clustering) of grouped data -- it is a generalization of the Chinese Restaurant Process. The CRF can be used to specify a nonparametric distribution on a mixture of mixtures: each grouping of data is a draw from a mixture model, where the mixture components are shared among different groups.
Context
This concept has the prerequisites:
- latent Dirichlet allocation (An important motivator for the CRF is Bayesian topic models.)
- CRP clustering (CRF topic models can be seen as a hierarchical extension of CRP clustering.)
Core resources (we're sorry, we haven't finished tracking down resources for this concept yet)
Supplemental resources (the following are optional, but you may find them useful)
-Free-
→ Princeton BNP Class: Lecture Notes 7 (2007)
→ Bayesian Nonparametrics (2011)
See also
- The CRF can be interpreted as a hierarchical Dirichlet process .