An Innovative BERT-Based Readability Model

INIS

• levels 100%
• transformers 100%
• prediction 83%
• china 50%
• performance 16%
• learning 16%
• information 16%
• assessments 16%
• errors 16%
• capture 16%
• processing 16%
• accuracy 16%
• classification 16%
• benchmarks 16%
• retention 16%

Computer Science

• Bidirectional Encoder Representations From Transformers 100%
• Level Representation 100%
• Prediction Model 80%
• Syntactics 20%
• Classification Models 20%
• Benchmark Task 20%
• Preprocessing 20%
• Transformer Model 20%