Unigram-Normalized Perplexity as a Language Model Performance Measure with Different Vocabulary Sizes.
Jihyeon RohSang-Hoon OhSoo-Young LeePublished in: CoRR (2020)
Keyphrases
- language model
- language modeling
- n gram
- similarity measure
- out of vocabulary
- retrieval model
- information retrieval
- document retrieval
- language modelling
- probabilistic model
- query expansion
- mixture model
- document length
- vector space model
- speech recognition
- test collection
- smoothing methods
- statistical language models
- word error rate
- context sensitive
- ad hoc information retrieval
- translation model
- document ranking
- language models for information retrieval
- distance measure
- information retrieval systems
- language model for information retrieval
- relevance model
- pseudo relevance feedback
- query terms