Pachinko allocation
Wei Li, Andrew McCallum · International Conference on Machine Learning · 2006
Latent Dirichlet allocation (LDA) and other related topic models are increasingly popular tools for summarization and manifold discovery in discrete data. However, LDA does not capture correlations between topics. In this paper, we introduce the pachinko allocation model (PAM), which captures arbitrary, nested, and possibly sparse correlations between topics using a directed acyclic graph (DAG). The leaves of the DAG represent individual words in the vocabulary, while each interior node represents a correlation among its children, which may be words or other interior nodes (topics). PAM provides a flexible alternative to recent work by Blei and Lafferty (2006), which captures correlations only between pairs of topics. Using text data from newsgroups, historic NIPS proceedings and other research paper corpora, we show improved performance of PAM in document classification, likelihood of held-out data, the ability to support finer-grained topics, and topical keyword coherence.
Cited by 609 works. doi:10.1145/1143844.1143917
Authors
References
- Latent dirichlet allocation
- Finding scientific topics
- Hierarchical Dirichlet Processes
- The author-topic model for authors and documents
- Marginal Likelihood from the Gibbs Output
- Approximate Bayesian Inference with the Weighted Likelihood Bootstrap
- Integrating Topics and Syntax
- Correlated Topic Models
- Hierarchical Topic Models and the Nested Chinese Restaurant Process
- Monte Carlo Methods of Inference for Implicit Statistical Models