Bayesian Agglomerative Clustering with Coalescents

Abstract

We introduce a new Bayesian model for hierarchical clustering based on a prior over trees called Kingman's coalescent. We develop novel greedy and sequential Monte Carlo inferences which operate in a bottom-up agglomerative fashion. We show experimentally the superiority of our algorithms over others, and demonstrate our approach in document clustering and phylolinguistics.

NIPS 2008

References in corpus (1)

• A Bayesian Model for Discovering Typological Implications

Cited by in corpus (2)

• The Infinite Hierarchical Factor Regression Model
• Bayesian Multitask Learning with Latent Hierarchies