Quasi Bidirectional Encoder Representations from Transformers for Word Sense Disambiguation.
Michele BevilacquaRoberto NavigliPublished in: RANLP (2019)
Keyphrases
- word sense disambiguation
- wordnet
- natural language processing
- word sense
- feature generation
- machine translation
- highly ambiguous
- semantic similarity
- bit rate
- information extraction
- semantic relatedness
- part of speech
- linguistic knowledge
- lexical information
- lexical knowledge
- natural language
- machine learning methods
- feature selection
- data mining
- knowledge base
- wide coverage