HiTR: Hierarchical Topic Model Re-Estimation for Measuring Topical Diversity of Documents.
Hosein AzarbonyadMostafa DehghaniTom KenterMaarten MarxJaap KampsMaarten de RijkePublished in: IEEE Trans. Knowl. Data Eng. (2019)
Keyphrases
- topic models
- topic modeling
- text documents
- latent topics
- latent dirichlet allocation
- pitman yor process
- probabilistic topic models
- topic discovery
- text mining
- lda model
- author topic model
- co occurrence
- text corpora
- relevance model
- generative model
- probabilistic model
- latent semantic analysis
- statistical topic models
- baseline models
- text analysis
- latent variables
- document collections
- document clustering
- information retrieval systems
- text data
- news articles
- language modeling framework
- generative process
- keywords
- information retrieval
- document retrieval
- word pairs
- probabilistic latent semantic analysis
- gibbs sampling
- document classification
- text streams
- variational inference
- parameter estimation
- latent topic model
- natural language processing