Korean Semantic Role Labeling with Bidirectional Encoder Representations from Transformers and Simple Semantic Information

State-of-the-art semantic role labeling (SRL) performance has been achieved using neural network models by incorporating syntactic feature information such as dependency trees. In recent years, breakthroughs end-to-end have resulted in a state-of-the-art SRL even without features. With the advent of language model called bidirectional encoder representations from transformers (BERT), another breakthrough was witnessed. Even though each word constituting sentence is important determining meaning word, previous studies regarding method did not utilize information. this study, we propose BERT-based that uses simple To obtain latter, used PropBank, which described relational between predicates and arguments. addition, text-originated obtained training text data utilized. Our proposed results on both Korean PropBank CoNLL-2009 English benchmarks.