Improving Japanese semantic-role-labeling performance with transfer learning as case for limited resources of tagged corpora on aggregated language.
Takuya OkamuraKoichi TakeuchiYasuhiro IshiharaMasahiro TaguchiYoshihiko InadaMasaya IizukaTatsuhiko AboHitoshi UedaPublished in: PACLIC (2018)
Keyphrases
- limited resources
- transfer learning
- semantic role labeling
- syntactic parsing
- dependency parsing
- knowledge transfer
- learning tasks
- natural language processing
- collaborative filtering
- labeled data
- active learning
- cross domain
- semi supervised learning
- reinforcement learning
- text mining
- machine learning
- text classification
- learning algorithm
- domain adaptation
- named entity recognition
- natural language
- artificial intelligence
- cross lingual
- word sense disambiguation
- machine learning algorithms
- semantic analysis
- text categorization
- pairwise
- relation extraction
- training data