Building topic mixture language models using the document soft classification notion of topic models.
Shuanhu BaiCheung-Chi LeungChien-Lin HuangBin MaHaizhou LiPublished in: ISCSLP (2010)
Keyphrases
- topic models
- language model
- relevance model
- probabilistic model
- language modeling framework
- latent topics
- topic modeling
- text documents
- latent dirichlet allocation
- topic discovery
- mixture model
- lda model
- document ranking
- language modeling
- document retrieval
- document classification
- text mining
- n gram
- text classification
- information retrieval
- document representation
- generative model
- vector space model
- query terms
- document level
- statistical topic models
- co occurrence
- query expansion
- image classification
- retrieval model
- expectation maximization
- pseudo feedback
- machine learning
- term frequency
- pseudo relevance feedback
- wordnet
- text corpora
- query specific
- smoothing methods
- text categorization
- class labels
- context dependent
- document clustering
- information extraction
- data mining