A BERT-Based Automatic Scoring Model of Korean Language Learners' Essay
0
조회
0
다운로드
5
인용
0
참고문헌
키워드
초록
This research applies a pre-trained bidirectional encoder representations from transformers (BERT) handwritingrecognition model to predict foreign Korean-language learners’ writing scores. A corpus of 586 answersto midterm and final exams written by foreign learners at the Intermediate 1 level was acquired and used forpre-training, resulting in consistent performance, even with small datasets. The test data were pre-processedand fine-tuned, and the results were calculated in the form of a score prediction. The difference between theprediction and actual score was then calculated. An accuracy of 95.8% was demonstrated, indicating that theprediction results were strong overall; hence, the tool is suitable for the automatic scoring of Korean writtentest answers, including grammatical errors, written by foreigners. These results are particularly meaningful inthat the data included written language text produced by foreign learners, not native speakers.
Abstract
This research applies a pre-trained bidirectional encoder representations from transformers (BERT) handwritingrecognition model to predict foreign Korean-language learners’ writing scores. A corpus of 586 answersto midterm and final exams written by foreign learners at the Intermediate 1 level was acquired and used forpre-training, resulting in consistent performance, even with small datasets. The test data were pre-processedand fine-tuned, and the results were calculated in the form of a score prediction. The difference between theprediction and actual score was then calculated. An accuracy of 95.8% was demonstrated, indicating that theprediction results were strong overall; hence, the tool is suitable for the automatic scoring of Korean writtentest answers, including grammatical errors, written by foreigners. These results are particularly meaningful inthat the data included written language text produced by foreign learners, not native speakers.
참고문헌 (13)
- 안수현, "한국어능력시험(TOPIK) 쓰기 평가의 채점 특성 연구", 한국어교육, 28(1), pp.173~196 (2017), 국제한국어교육학회[DOI]
- 황상흠, "한국어 기술문서 분석을 위한 BERT 기반의 분류모델", 한국전자거래학회지, 25(1), pp.203~214 (2020), 한국전자거래학회[DOI]
- 민재옥, "BERT를 이용한 한국어 특허상담 기계독해", 정보처리학회 논문지, 9(4), pp.145~152 (2020), 한국정보처리학회[DOI]
- 이치훈, "사전 학습된 한국어 BERT의 전이학습을 통한 한국어 기계독해 성능개선에 관한 연구", 한국IT서비스학회지, 19(5), pp.83~91 (2020), 한국IT서비스학회[DOI]
- K. Jiang, "Natural language processing and its applications in machine translation: a diachronic review", Proceedings of 2020 IEEE 3rd International Conference of Safe Production and Informatization (IICSPI), pp.210~214 (2020)
- J. Devlin, "BERT : pre-training of deep bidirectional transformers for language understanding", Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), pp.4171~4186 (2019)
- D. Alikaniotis, "Automatic text scoring using neural networks", Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (ACL), pp.715~725 (2016)
- J. E. Kim, "Automatic scoring system for short descriptive answer written in Korean using lexico-semantic pattern", Soft Computing, 22(13), pp.4241~4249 (2018)
- National Institute of the Korean Language, "Application Research of Korean Language Curriculum" (2017), National Institute of the Korean Language
- 이정희, "한국어 오류 판정과 분류 방법에 관한 연구", 한국어교육, 13(1), pp.175~197 (2002), 국제한국어교육학회
- "BERT with SentencePiece for Korean Text" (2020)[URL]
- H. Lee, "KoreALBERT : pretraining a Lite BERT model for Korean language understanding", Proceedings of 2020 25th International Conference on Pattern Recognition (ICPR), pp.5551~5557 (2021)
- "Test of Proficiency of Korea (TOPIC)"[URL]