DSL Shared Task 2016: Perfect Is The Enemy of Good Language Discrimination Through Expectation-Maximization and Chunk-based Language Model.
Ondrej HermanVit SuchomelVít BaisaPavel RychlýPublished in: VarDial@COLING (2016)
Keyphrases
- language model
- expectation maximization
- probabilistic model
- mixture model
- language modeling
- em algorithm
- n gram
- document retrieval
- generative model
- query expansion
- information retrieval
- speech recognition
- test collection
- language modelling
- retrieval model
- maximum likelihood
- document ranking
- programming language
- statistical language models
- ad hoc information retrieval
- context sensitive
- translation model
- smoothing methods
- natural language
- maximum a posteriori
- pseudo relevance feedback
- language model for information retrieval
- context dependent
- vector space model
- query terms
- parameter estimation
- bayesian networks
- target language
- relevance model
- coreference resolution
- image segmentation
- language models for information retrieval
- feature selection