Sometimes Average is Best: The Importance of Averaging for Prediction using MCMC Inference in Topic Modeling.
Viet-An NguyenJordan L. Boyd-GraberPhilip ResnikPublished in: EMNLP (2014)
Keyphrases
- topic modeling
- topic models
- gibbs sampling
- latent dirichlet allocation
- markov chain monte carlo
- generative model
- modeling framework
- bayesian inference
- topic extraction
- text mining
- text corpora
- text classification
- latent topics
- particle filter
- collaborative filtering
- probabilistic latent semantic analysis
- approximate inference
- bayesian networks
- probabilistic model
- probabilistic inference
- training data
- search engine
- belief propagation
- exact inference
- databases
- natural language
- pattern recognition
- neural network