Journal of The Korean Association of Information Education (정보교육학회논문지)

Korea Association of Information Education (한국정보교육학회)
권호 표지
DOI QR코드

DOI QR Code

Automatic Evaluation of Elementary School English Writing Based on Recurrent Neural Network Language Model

순환 신경망 기반 언어 모델을 활용한 초등 영어 글쓰기 자동 평가

  • Park, Youngki (Department of Computer Education, Chuncheon National University of Education)
  • Received : 2016.10.24
  • Accepted : 2017.01.31
  • Published : 2017.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

We often use spellcheckers in order to correct the syntactic errors in our documents. However, these computer programs are not enough for elementary school students, because their sentences are not smooth even after correcting the syntactic errors in many cases. In this paper, we introduce an automated method for evaluating the smoothness of two synonymous sentences. This method uses a recurrent neural network to solve the problem of long-term dependencies and exploits subwords to cope with the rare word problem. We trained the recurrent neural network language model based on a monolingual corpus of about two million English sentences. In our experiments, the trained model successfully selected the more smooth sentences for all of nine types of test set. We expect that our approach will help in elementary school writing after being implemented as an application for smart devices.

작성된 문서의 문법적 오류 교정을 할 때 맞춤법 검사기를 사용하는 것이 일반적이다. 그러나 초등학생들이 작성한 글 중에는 문법적으로는 옳더라도 자연스럽지 않은 문장이 있을 수 있다. 본 논문에서는 동일한 의미를 가진 2개의 문장이 주어졌을 때, 어떤 것이 더 자연스러운 문장인지 자동 판별할 수 있는 방법을 소개한다. 이 방법은 순환 신경망(recurrent neural network)을 이용하여 장기 의존성(long-term dependencies) 문제를 해결하고, 보조 단어(subword)를 사용하여 희소 단어(rare word) 문제를 해결한다. 약 200만 문장의 단일어 코퍼스를 통해 순환 신경망 기반 언어 모델을 학습하였다. 그 결과, 초등학생들이 주로 틀리는 표현들과 그에 대응하는 올바른 표현을 입력으로 주었을 때, 모든 경우에 대해 자연스러운 표현을 자동으로 선별할 수 있었다. 본 소프트웨어가 스마트 기기에 사용될 수 있는 형태로 구현된다면 실제 초등학교 현장에서 활용 가능할 것으로 기대된다.

Keywords

References

  1. Bahdanau, D., Cho, K. & Bengio, Y. (2015). Neural Machine Translation By Jointly Learning to Align and Translate. In Proceedings of the International Conference on Learning Representations, 1-15.
  2. Cho, K. (2015). Deep Learning for Machine Translation. DL4MT Winter School.
  3. Gonzalez, Y., Saenz, L., Bermeo, J. & Chaves, A. (2013). The Role of Collaborative Work in the Development of Elementary Students’ Writing Skills. Profile Issues in Teacher’ Professsional Development, 15(1), 11-25.
  4. Graves, A. (2013). Generating Sequences with Recurrent Neural Networks. arXiv:1308.0850.
  5. Hochreiter, S. & Schmidhuber, J. (1997). Long Short-term Memory. Neural Computation, 9(8), 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  6. In, J. & Han, J. (2016). The Prosodic Changes of Korean English Learners in Robot Assisted Learning. Journal of the Korean Association of Information Education, 20(4), 323-332.
  7. Jang, S., Chun, S. (2016). A Study on Teachers’ Use of Applications in Teaching-Learning Activities. Journal of the Korea Association of Information Education, 20(1), 1-12. https://doi.org/10.14352/jkaie.2016.20.1.1
  8. Jeon, S. & Kim, H. (2016). A Study on Improvement of Web-based Diagnosis-Supplement System for Basic Academic Skills, 20(5), 487-498.
  9. Kim, C. (2014). A Study on the Educational Use of Tiny PC in an Elementary School. Journal of the Korean Assocation of Information Education, 18(1). 101-110. https://doi.org/10.14352/jkaie.2014.18.1.101
  10. Koehn, P. (2005). Europarl: A Parallel Corpus for Statistical Machine Translation. MT Summit.
  11. Lee, J. & Yoo, S. (2014). A Data Logging Smart r-Learning Effect on Students’ Logical Thinking. Journal of the Korean Association of Information Education, 18(1), 25-33. https://doi.org/10.14352/jkaie.2014.18.1.25
  12. Lee, M. & Ham, S. (2015). The Development and Effectiveness of the Smart System for Supporting Instructional Materials. Journal of the Korean Association of Information Education, 19(4), 399-408. https://doi.org/10.14352/jkaie.2015.19.4.399
  13. Lee, Y., Oh, D. & Park, S. (2015). The Web-Based Interface Design for University Students’ Activity-Oriented Career Education. Journal of the Korean Association of Information Education, 19(3), 345-360. https://doi.org/10.14352/jkaie.2015.19.3.345
  14. Luong, M., Pham, H. & Manning, C. D. (2015). Effective Approaches to Attention-based Neural Machine Translation. arXiv:1508.04025.
  15. Park, S., Han, K., Lee, D., Shin, B. & Lee, J. (2015). Designing and Developing ICT Contents of Mathematics and Science in Agricultural, Mountain and Fishing Villages. Journal of the Korean Association of Information Education, 19(2), 215-224. https://doi.org/10.14352/jkaie.2015.19.2.215
  16. Park, Y. (2016). Automatic Generation of Multiple-Choice Questions Based on Statistical Language Model. Journal of the Korean Association of Information Education, 20(2), 197-206. https://doi.org/10.14352/jkaie.20.2.197
  17. Park. Y. (2016). Automatic Selection of Similar Sentences for Teaching Writing in Elementary School. Journal of the Korean Assocation of Information Education, 20(4), 333-340.
  18. Pham, T. (2012). A Study on Teaching and Learning Korean Grammars Method Based on Paraphrasing Activities. Master's Thesis, Seoul National University.
  19. Ryu, M. & Han, S. (2016). Development of Smart Application for English Speaking. Journal of the Korean Association of Information Education, 20(4), 367-374.
  20. Sennrich, R., Haddow, B. & Birch, A. (2016). Neural Machine Translation of Rare Words with Subword Units. In Proceedings of the Annual Meeting of the Association for Computational Linguistics, arXiv:1508.07909.
  21. Suh, S. & Goh, Y. (2016). Study on School Teachers’ Perception of and Usage of SMART Education. Journal of the Korean Association of Information Education, 20(2), 139-150. https://doi.org/10.14352/jkaie.20.2.139
  22. Thornbury, S. (2000). How to Teach Grammar. Longman.
  23. Xu, K., Ba, J. L., Kiros, R., Cho, K., Courville, A., Salakhutdinov, R., Zemel, R. S. & Bengio, Y. (2015). Show, Attend and Tell: Neural Image Caption Generation with Visual Attention. arXiv:1502.03044.