Exploring word embeddings and phonological similarity for the unsupervised correction of language learner errors.
Ildikó PilánElena VolodinaPublished in: LaTeCH@COLING (2018)
Keyphrases
- language learners
- language learning
- pointwise mutual information
- syntactic categories
- word recognition
- error analysis
- lexical information
- chinese text retrieval
- error detection
- linguistic knowledge
- distance measure
- native speakers
- word similarity
- programming language
- text to speech synthesis
- similarity measure
- word sense induction
- learning environment
- sentence similarity
- unsupervised learning
- foreign language
- error detection and correction
- language specific
- learning process
- english text
- target language
- co occurrence
- n gram
- word segmentation
- euclidean distance
- error correction
- language processing
- information extraction
- pos taggers
- word sense disambiguation
- pos tagging
- word order
- lexical semantics
- dimensionality reduction
- machine translation
- foreign language learning
- binary codes
- semi supervised
- learning from positive data
- word sense
- machine translation system