Improving Japanese semantic-role-labeling performance with transfer learning as case for limited resources of tagged corpora on aggregated language

Arts & Humanities

• English People 18%
• Labeling 80%
• Language 34%
• Morpheme 27%
• Neural Networks 61%
• Performance 29%
• Resources 48%

Social Sciences

• CNN 36%
• learning 26%
• neural network 61%
• performance 35%
• resources 33%
• semantics 61%

Engineering & Materials Science

• Labeling 75%
• Neural networks 31%
• Transfer learning 91%