Revisiting CNN for Highly Inflected Bengali and Hindi Language Modeling.
Chowdhury Rafeed RahmanMd. Hasibur RahmanMohammad RafsanSamiha ZakirMohammed Eunus AliRafsanjani MuhammodPublished in: CoRR (2021)
Keyphrases
- language modeling
- cross lingual
- indian languages
- language model
- named entity recognition
- statistical machine translation
- information retrieval
- comparable corpora
- query expansion
- retrieval model
- n gram
- probabilistic model
- cross language
- language independent
- information extraction
- named entities
- translation model
- machine translation
- statistical language models
- word segmentation
- language identification
- text classification
- improvements in retrieval effectiveness
- statistical language modeling
- document images
- machine translation system
- smoothing methods
- semi supervised
- relevance model
- document retrieval
- conditional random fields
- test collection
- mixture model
- high dimensional
- search engine