Inducing Embeddings for Rare Words through Morphological Decomposition, Stemming and Bidirectional Translation.
Xiaotao LiShujuan YouWai ChenPublished in: ICMLA (2019)
Keyphrases
- word forms
- n gram
- language independent
- english words
- machine translation
- syntactic categories
- stop words
- multiword
- character n grams
- image processing
- translation model
- decomposition method
- parallel texts
- out of vocabulary
- keywords
- mathematical morphology
- manifold learning
- parallel corpora
- compound words
- shape decomposition
- word level
- word segmentation
- word alignment
- parallel corpus
- low dimensional
- multiscale
- related words
- euclidean space
- text classification
- structuring elements
- dimensionality reduction
- cross language information retrieval
- morphological analysis
- information retrieval
- language model
- preprocessing
- cross lingual