Document Filtering Based on Spectral Clustering for Speech Recognition Language Model.
Shinya TakahashiTsuyoshi MorimotoNaoyuki TsurutaPublished in: IMECS (2007)
Keyphrases
- speech recognition
- language model
- spectral clustering
- document retrieval
- document representation
- query terms
- information retrieval
- vector space model
- language modeling
- n gram
- clustering method
- data clustering
- probabilistic model
- clustering algorithm
- pairwise
- retrieval model
- relevance model
- automatic speech recognition
- test collection
- speech signal
- query expansion
- mixture model
- document clustering
- k means
- relevant documents
- information retrieval systems
- retrieval systems
- document collections
- web documents
- image segmentation
- tf idf
- clustering quality
- word error rate
- handwriting recognition
- document images
- pattern recognition
- average precision
- term frequency
- co occurrence
- hidden markov models
- bayesian networks
- similarity measure
- translation model
- out of vocabulary
- data mining