Japanese Spontaneous Spoken Document Retrieval Using NMF-Based Topic Models.
Xinhui HuHideki KashiokaRyosuke IsotaniSatoshi NakamuraPublished in: AIRS (2009)
Keyphrases
- topic models
- spoken document retrieval
- negative matrix factorization
- probabilistic latent semantic analysis
- topic modeling
- latent dirichlet allocation
- information retrieval
- cross language
- text mining
- text documents
- broadcast news
- test collection
- document clustering
- latent topics
- probabilistic model
- generative model
- co occurrence
- text classification
- maximum likelihood
- information extraction
- text corpora
- search engine