The nested Chinese restaurant process and Bayesian nonparametric inference of topic hierarchies
From MaRDI portal
Abstract: We present the nested Chinese restaurant process (nCRP), a stochastic process which assigns probability distributions to infinitely-deep, infinitely-branching trees. We show how this stochastic process can be used as a prior distribution in a Bayesian nonparametric model of document collections. Specifically, we present an application to information retrieval in which documents are modeled as paths down a random tree, and the preferential attachment dynamics of the nCRP leads to clustering of documents according to sharing of topics at multiple levels of abstraction. Given a corpus of documents, a posterior inference algorithm finds an approximation to a posterior distribution over trees, topics and allocations of words to levels of the tree. We demonstrate this algorithm on collections of scientific abstracts from several journals. This model exemplifies a recent trend in statistical machine learning--the use of Bayesian nonparametric methods to infer distributions on flexible data structures.
Recommendations
- Hierarchical Dirichlet Processes
- Hierarchical topic modeling with nested hierarchical Dirichlet process
- Non-parametric estimation of topic hierarchies from texts with hierarchical Dirichlet processes
- Distance dependent chinese restaurant processes
- Latent nested nonparametric priors (with discussion)
Cited in
(34)- Regenerative tree growth: binary self-similar continuum random trees and Poisson-Dirichlet compositions
- Hierarchical topic modeling with nested hierarchical Dirichlet process
- A representation of exchangeable hierarchies by sampling from random real trees
- Latent tree models for hierarchical topic detection
- Sampling and learning Mallows and generalized Mallows models under the Cayley distance
- Graph-induced restricted Boltzmann machines for document modeling
- Bayesian nonparametric modeling and the ubiquitous Ewens sampling formula
- Exchangeable hierarchies and mass-structure of weighted real trees
- Concentration in the generalized Chinese restaurant process
- Local and global topics in text modeling of web pages nested in web sites
- Fair division in the presence of externalities
- Exceptional spatio-temporal behavior mining through Bayesian non-parametric modeling
- Latent nested nonparametric priors (with discussion)
- Cooperative hierarchical Dirichlet processes: superposition vs. maximization
- Nonparametric Bayesian topic modelling with the hierarchical Pitman-Yor processes
- Hierarchical Dirichlet Processes
- Statistical topic models for multi-label document classification
- Sampling, denoising and compression of matrices by coherent matrix organization
- A multiscale community blockmodel for network exploration
- Text mining in computational advertising
- Distance dependent chinese restaurant processes
- Non-parametric estimation of topic hierarchies from texts with hierarchical Dirichlet processes
- Graph Neural Networks for Natural Language Processing: A Survey
- Cluster and feature modeling from combinatorial stochastic processes
- Structured sparsity through convex optimization
- A down‐up chain with persistent labels on multifurcating trees
- Bayesian Tree-Structured Two-Level Clustering for Nested Data Analysis
- Unsupervised learning for medical data: a review of probabilistic factorization methods
- A Bayesian nonparametric model for multi-label learning
- Nested Dirichlet models for unsupervised attack pattern detection in honeypot data
- A multifacet hierarchical sentiment-topic model with application to multi-brand online review analysis
- Independent finite approximations for Bayesian nonparametric inference
- Learning topic hierarchies by tree-directed latent variable models
- Up-down ordered Chinese restaurant processes with two-sided immigration, emigration and diffusion limits
This page was built for publication: The nested Chinese restaurant process and Bayesian nonparametric inference of topic hierarchies
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3578188)
