DOI QR코드

DOI QR Code

The Study on Possibility of Applying Word-Level Word Embedding Model of Literature Related to NOS -Focus on Qualitative Performance Evaluation-

과학의 본성 관련 문헌들의 단어수준 워드임베딩 모델 적용 가능성 탐색 -정성적 성능 평가를 중심으로-

  • Received : 2021.01.11
  • Accepted : 2022.04.18
  • Published : 2022.04.30

Abstract

The purpose of this study is to look qualitatively into how efficiently and reasonably a computer can learn themes related to the Nature of Science (NOS). In this regard, a corpus has been constructed focusing on literature (920 abstracts) related to NOS, and factors of the optimized Word2Vec (CBOW, Skip-gram) were confirmed. According to the four dimensions (Inquiry, Thinking, Knowledge and STS) of NOS, the comparative evaluation on the word-level word embedding was conducted. As a result of the study, according to the previous studies and the pre-evaluation on performance, the CBOW model was determined to be 200 for the dimension, five for the number of threads, ten for the minimum frequency, 100 for the number of repetition and one for the context range. And the Skip-gram model was determined to be 200 for the number of dimension, five for the number of threads, ten for the minimum frequency, 200 for the number of repetition and three for the context range. The Skip-gram had better performance in the dimension of Inquiry in terms of types of words with high similarity by model, which was checked by applying it to the four dimensions of NOS. In the dimensions of Thinking and Knowledge, there was no difference in the embedding performance of both models, but in case of words with high similarity for each model, they are sharing the name of a reciprocal domain so it seems that it is required to apply other models additionally in order to learn properly. It was evaluated that the dimension of STS also had the embedding performance that was not sufficient to look into comprehensive STS elements, while listing words related to solution of problems excessively. It is expected that overall implications on models available for science education and utilization of artificial intelligence could be given by making a computer learn themes related to NOS through this study.

본 연구의 목적은 NOS 관련 주제를 대상으로 컴퓨터가 얼마나 효율적이고 타당하게 학습할 수 있는지에 대하여 정성적으로 탐색하고자 한 연구이다. 이를 위해 NOS와 관련되는 문헌(논문초록 920편)을 중심으로 말뭉치를 구성하였으며, 최적화된 Word2Vec (CBOW, Skip-gram)모델의 인자를 확인하였다. 그리고 NOS의 4가지 영역(Inquiry, Thinking, Knowledge, STS)에 따라 단어수준 워드임베딩 모델 비교평가를 수행하였다. 연구 결과, 선행연구와 사전 성능 평가에 따라 CBOW 모델은 차원 200, 스레드 수 5, 최소빈도수 10, 반복횟수 100, 맥락범위 1로 결정되었으며, Skip-gram 모델은 차원수 200, 스레드 수 5, 최소빈도수 10, 반복횟수 200, 맥락범위 3으로 결정되었다. NOS의 4가지 영역에 적용하여 확인한 모델별 유사도가 높은 단어의 종류는 Skip-gram 모델이 Inquiry 영역에서 성능이 좋았다. Thinking 및 Knowledge 영역에서는 두 모델별 임베딩 성능 차이는 나타나지 않았으나, 각 모델별 유사도가 높은 단어의 경우 상호 영역 명을 공유하고 있어 제대로 된 학습을 하기 위해 다른 모델의 추가 적용이 필요해 보였다. STS 영역에서도 지나치게 문제 해결과 관련된 단어를 나열하면서 포괄적인 STS 요소를 탐색하기에 부족한 임베딩 성능을 지닌 것으로 평가되었다. 본 연구를 통해 NOS 관련 주제를 컴퓨터에게 학습시켜 과학교육에 활용할 수 있는 모델과 인공지능 활용에 대한 전반적인 시사점을 줄 수 있을 것으로 기대된다.

Keywords

References

  1. Abd-El-Khalick, F. (2005). Developing deeper understandings of nature of science: The impact of a philosophy of science course on preservice science teachers' views and instructional planning. International Journal of Science Education, 27(1), 15-42. https://doi.org/10.1080/09500690410001673810
  2. Abell, S., Martini, M., & George, M. (2001). 'That's what scientists have to do': Preservice elementary teachers' conceptions of the nature of science during a moon investigation. Journal of Research in Science Teaching, 23(11), 1095-1109.
  3. Ackerson, V. L., Morrison, J. A., & McDuffie, A. R. (2006). One course is not enough: Preservice elementary teachers' retention of improved views of nature of science. Journal of Research in Science Teaching, 43, 194-213. https://doi.org/10.1002/tea.20099
  4. Akerson, V. L., Buzzelli, C., & Donnelly, L. A. (2010). On the nature of teaching nature of science: Preservice early childhood teachers' instruction in preschool and elementary settings. Journal of Research in Science Teaching, 47, 213-233. https://doi.org/10.1002/tea.20323
  5. American Association for the Advancement of Science [AAAS]. (1990). Science for all Americans. New York, NY: Oxford University Press.
  6. Aoun, J. E. (2017). ROBOT-PROOF: Higher education in the age of artificial intelligence. Cambridge, MA: MIT Press.
  7. Bartholomew, H., Osborne, J., & Ratcliffe, M. (2004). Teaching pupils "ideas-aboutscience": Five dimensions of effective practice. Science Education, 88, 655-682. https://doi.org/10.1002/sce.10136
  8. Bayir, E., Cakici, Y., & Ertas, O. (2014). Exploring natural and social scientists' views of nature of science. International Journal of Science Education, 36(8), 1286-1312. https://doi.org/10.1080/09500693.2013.860496
  9. Chiappetta, E. L., & Fillman, D. A. (2005). Analysis of five high school biology textbooks used in the United States for inclusion of the nature of science. Paper presented at the National Association for Research in Science Teaching meeting. Dallas, TX.
  10. Chiappetta, E. L., Fillman, D. A., & Sethna, G. H. (1991). A method to quantify major themes of scientific literacy in science textbooks. Journal of Research in Science Teaching, 28, 713-725. https://doi.org/10.1002/tea.3660280808
  11. Chiappetta, E. L., Sethna, G. H., & Fillman, D. A. (1991). A qualitative analysis of high school chemistry textbooks for scientific literacy themes and expository learning aids. Journal of Research in Science Teaching, 28, 936-951.
  12. Choi, S., Matteson, A. S., & Lim, H. (2018). Utilizing local bilingual embeddings on Korean-English law data. Journal of the Korea Convergence Society, 9(10), 45-53. https://doi.org/10.15207/JKCS.2018.9.10.045
  13. Cobern, W. W., & Loving, C. C. (2002). Investigation of preservice elementary teachers' thinking about science. Journal of Research in Science Teaching, 39(10), 1016-1-31. https://doi.org/10.1002/tea.10052
  14. Collette, A., & Chiappetta, L. E. (1984). Science instruction in the middle and secondary schools. St. Louis, MO: Times Millor/Mosby.
  15. Duschl, R. (2000). Making the nature of science explicit. In R. Millar, J. Leach, & J. Osborne (Eds.), Improving science education: The contribution of research (pp. 187-206). Philadelphia, PA: Open University Press.
  16. Harding, P., & Hare, W. (2000). Portraying science accurately in classrooms: Emphasizing open-mindedness rather than relativism. Journal of Research in Science Teaching, 37(3), 225-236. https://doi.org/10.1002/(SICI)1098-2736(200003)37:3<225::AID-TEA1>3.0.CO;2-G
  17. Kang, H., & Yang, J. (2019). Optimization of Word2vec models for Korean word embeddings. Journal of Digital Contents Society, 20(4), 825-833. https://doi.org/10.9728/dcs.2019.20.4.825
  18. Kim, G., Kang, G., Son, M., Lee C., Hong, S., & Kim, S. (2020). A big-data analysis of issues on North Korea and media agenda setting functions: Applying topic modeling and word-embedding methods. Peace Studies, 28(1), 287-332. https://doi.org/10.21051/ps.2020.04.28.1.287
  19. Kim, H. (2021). The artificial intelligence era and science education -With a focus on the autonomy and relatedness of artificial intelligence-. The Journal of Yeolin Education, 29(6), 1-23
  20. Kim, Y., Lee, B., & Ha, Y. (2015). Scientific reasoning and nature of science found in the history of atomic model development -College science for scientific literacy. Korean Journal of General Education, 9(2), 347-376.
  21. Korawit, O., & Wu, Y. (2019). Word sense disambiguation using cosine similarity collaborates with Word2vec and WordNet. Future Internet, 11(5), 114. https://doi.org/10.3390/fi11050114
  22. Kottur, S., Vedantam, R., Moura, J. M. F., & Parikh, D. (2015). Visual word2vec (vis-w2v): Learning visually grounded word embeddings using abstract scenes. CoRR, abs/1511.07067.
  23. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems, 1097-1105.
  24. Lederman, N. G. (1992). Students' and teachers' conceptions of the nature of science: A review of the research. Journal of Research in Science Teaching, 29(4), 331-359. https://doi.org/10.1002/tea.3660290404
  25. Lee, G. (2019). Korean Embedding. Seoul: Acon.
  26. Lee, Y. (2013). A proposal of inclusive framework of the Nature of Science (NOS) based on the 4 themes of scientific literacy for K-12 school science. Journal of the Korean Association for Science Education, 33(3), 553-568. https://doi.org/10.14697/JKASE.2013.33.3.553
  27. Lee, Y. (2014a). Comparative analysis of the presentation of the Nature of Science (NOS) in Korea and US elementary science textbooks. Journal of the Korean Association for Science Education, 34(3), 207-212. https://doi.org/10.14697/JKASE.2014.34.3.0207
  28. Lee, Y. (2014b). What do scientists think about the Nature of Science? -Exploring views of the Nature of Science of Korean scientists related with life science area. Journal of the Korean Association for Science Education, 34(7), 677-691. https://doi.org/10.14697/JKASE.2014.34.7.0677
  29. Lee, Y., Son, Y., & Kim, K. (2014). Analysis of the presentation for the Nature of Science in elementary science textbooks using the four themes of scientific literacy. Journal of Korean Elementary Science Education, 33(2), 207-216. https://doi.org/10.15267/KESES.2014.33.2.207
  30. Sahlgren, M. (2008). The distributional hypothesis. Italian Journal of Disability Studies, 20, 33-53.
  31. McComas, W. F. (2005). Seeking NOS standards: What content consensus exists in popular books on the nature of science?. Paper presented at the National Association for Research in Science Teaching meeting. Dallas, TX.
  32. McDonald, C. V. (2010). The influence of explicit nature of science and argumentation instruction on preservice teachers' views of nature of science. Journal of Research in Science Teaching, 47, 1137-1164. https://doi.org/10.1002/tea.20377
  33. Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013). Efficient estimation of word representations in vector space. arXiv preprint arXiv: 1301.3781.
  34. Mikolov, T., Sutskever, I., Chen, K., Corrado, G., & Dean, J. (2013). Distributed representations of words and phases and their compositionality. Advances in neural information processing systems, 3111-3119.
  35. National Research Council [NRC]. (1996). National science education standards. Washington, DC: National Academy Press.
  36. National Research Council [NRC]. (2012). A framework for K-12 science education. Washington, DC: National Academy Press.
  37. Oliveira, A. W., Akerson, V. L., Colak, H., Pongsanon, K., & Genel, A. (2012). The implicit communication of nature of science and epistemology during inquiry discussion. Science Education, 96, 652-684. https://doi.org/10.1002/sce.21005
  38. Raschka, S., & Mirjalili, V. (2019). Python machine learning: Machine learning and deep learning with python, scikit-learn, and TensorFlow 2. Birmingham, England: Packt Publishing Ltd.
  39. Russell, S., & Bohannon, J. (2015). Artificial intelligence. Fears of an AI pioneer. Science, 349(6245), 252. https://doi.org/10.1126/science.349.6245.252
  40. Schwartz, R., & Lederman, N. (2008). What scientists say: Scientists' views of nature of science and relation to science context. International Journal of Science Education, 30(6), 727-771. https://doi.org/10.1080/09500690701225801
  41. Shin, D., & Kim, C. (2016). Query extension of retrieve system using Hangul word embedding and Apriori. Journal of Advanced navigation Technology, 20(6), 617-624. https://doi.org/10.12673/JANT.2016.20.6.617
  42. Shin, S., & Kim, K. (2016). Addressing the new user problem of recommender systems based on word embedding learning and skip-gram modelling. Journal of The Korea Society of Computer and Information, 21(7), 9-16. https://doi.org/10.9708/JKSCI.2016.21.7.009
  43. Shin, W., & Shin, D. (2020). A study oh the artificial intelligence in elementary science education. Journal of Korean Elementary Science Education, 39(1), 117-132. https://doi.org/10.15267/KESES.2020.39.1.117
  44. Taylor, A. R., Jones, M. G., Broadwell, B., & Oppewal, T. (2008). Creativity, inquiry, or accountability? Scientists' and teachers' perceptions of science education. Science Education, 92, 1058-1075. https://doi.org/10.1002/sce.20272
  45. Yun, E., & Park, Y. (2019). Qualitative performance evaluation of the word-embeddin model through learning science Textbook Corpus(K-STeC). New Physics: Sae Mulli, 69(10), 1038-1052. https://doi.org/10.3938/npsm.69.1038