Adapting Word Embeddings to New Languages with Morphological and Phonological Subword Representations.
Aditi ChaudharyChunting ZhouLori S. LevinGraham NeubigDavid R. MortensenJaime G. CarbonellPublished in: EMNLP (2018)
Keyphrases
- n gram
- language independent
- word forms
- word recognition
- spoken document retrieval
- language specific
- english text
- word clouds
- word segmentation
- expressive power
- compound words
- grammar induction
- out of vocabulary
- character n grams
- machine translation system
- multiscale
- statistical machine translation
- vector space
- syntactic categories
- co occurrence
- bilingual dictionaries
- broadcast news
- multiword
- target language
- text classification
- manifold learning
- language model
- cross lingual
- image processing
- word pairs
- intermediate representations
- word sense disambiguation
- structuring elements
- euclidean space
- indian languages
- speech recognizer
- speech recognition
- word order
- distance measure
- sentence level
- low dimensional