當今世界科技飛速發展，創意和溝通的管道可以有更廣泛的方法去傳播，傳統的面授教育也產生重大革新。在2019年12月，Covid-19突然爆發，迫使許多學生轉移到遠距教學模式。遠距教學中的非同步課程可以讓學生自行管理上課進度，但老師會不容易得知學生是否專心於課堂上。因此本研究想要為非同步遠距教學的老師，自動生成一系列以非同步遠距上課內容為主的問句，作為輔助教學的工具之一，而欲生成課程問句需透過句子選擇與問句建構兩步驟。
過往的問句生成的研究方法主要分為傳統基於規則的(Automatic Question Generation, AQG) 、基於神經網絡(Neural Question Generation, NQG)、以及近年來突然興起的預訓練問句生成。由於深度神經網路通常有大量參數，所以在沒有足夠訓練數據的情況下，容易擬合過度且泛化能力(generalization ability)較差，而預訓練問題生成可透過預訓練語言模型，僅需少量的樣本便可進行問題生成，且運算時間較短。然而現今的問句生成方法多為機器導向而非應用於教育領域，因此方法中會忽略句子選擇之步驟，但是生成課程問句時句子選擇應在其中扮演著重要的角色，必須選擇教材中對學習有幫助、值得出題的句子才有出題及學習的意義。因此本研究提出ADT-QG(Asynchronous Distance Teaching-Question Generation)模型，在模型架構中加入Sentences-BERT(SBERT)為了獲取相似度更高的句子來做問句生成，以及透過Text-to-Text Transformer (T5)模型來做答案-問句生成，並且配合Wiki語料庫生成選項，期望生成較流暢且符合教學內容之問句。
最後本研究實驗發現，與其他做相似度計算的模型相比，SBERT能選擇出更有資訊且可問的相關句子。而跟其他問題生成模型相比加入透過使用T5模型的ADT-QG模型能達到較佳的表現，且在人工評估上都有達到5分以上的表現，更優於過往模型。

With the rapid development of science and technology in today's world, the channels of creativity and communication can be spread in a wider range of methods, and traditional face-to-face education has also produced major innovations. In December 2019, the Covid-19 outbreak forced 1.57 billion students and teachers around the world to use distance learning. Therefore, this research wants to automatically generate a series of questions for asynchronous distance teaching as one of the tools to assist teaching.
In recent years, pre-trained language models can be used to generate questions, and only a small number of samples can be used to generate questions without retraining the model. However, in the field of education sentence selection should play an important role in generating curriculum questions. It is necessary to select sentences in the teaching material that are helpful to learning and worthy of questions. Therefore, this study proposes the ADT-QG(Asynchronous Distance Teaching-Question Generation)model, adding SBERT to the model architecture to obtain sentences with higher similarity, and training the question generation through the T5 model.
Finally, this research experiment found that compared with other models for similarity calculation, SBERT can select more informative and questionable related sentences. Compared with other question generation models, adding the ADT-QG model using the T5 model can achieve better performance, showing that the choice of the pre-training model really helps the model to grasp the direction of text generation.

Agarwal, M., & Mannem, P. (2011). Automatic Gap-Fill Question Generation from Text Books. Paper presented at the the 6th Workshop on Innovative Use of NLP for Building Educational Applications, Portland, Oregon.
Bahdanau, D., Cho, K., & Bengio, Y. (2014). Neural Machine Translation by Jointly Learning to Align and Translate. Paper presented at the arXiv preprint arXiv:1409.0473.
Belkin, M., Hsu, D., Ma, S., & Mandal, S. (2019). Reconciling modern machine-learning practice and the classical bias–variance trade-off. Proceedings of the National Academy of Sciences, 116(32), 15849-15854.
Burke, L. A., & Ray, R. (2008). Re-setting the concentration levels of students in higher education: an exploratory study. Teaching in Higher Education, 13(5), 571-582.
Chan, Y.-H., & Fan, Y.-C. (2019). A recurrent BERT-based model for question generation. Paper presented at the Proceedings of the 2nd Workshop on Machine Reading for Question Answering. Hong Kong, China.
Chen, G., Yang, J., & Gasevic, D. (2019). A Comparative Study on Question-Worthy Sentence Selection Strategies for Educational Question Generation. Paper presented at the the International Conference on Artificial Intelligence in Education, Chicago, USA.
Cheng, J., & Lapata, M. (Writers). (2016). Neural Summarization by Extracting Sentences and Words. In the 54th Annual Meeting of the Association for Computational Linguistics. Berlin, Germany
Cho, K., Van Merriënboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014). Learning Phrase Representations Using RNN Encoder-Decoder for Statistical Machine Translation. arXiv preprint arXiv:1406.1078.
d'Inverno, R., Davis, H., & White, S. (2003). Using a personal response system for promoting student interaction. Teaching Mathematics and Its Applications: International Journal of the IMA, 22(4), 163-169.
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
Dhawan, S. (2020). Online learning: A panacea in the time of COVID-19 crisis. Journal of Educational Technology Systems, 49(1), 5-22.
Dietrich, N., Kentheswaran, K., Ahmadi, A., Teychené, J., Bessière, Y., Alfenore, S., . . . Guigui, C. (2020). Attempts, successes, and failures of distance learning in the time of COVID-19. Journal of Chemical Education, 97(9), 2448-2457.
Du, X., Shao, J., & Cardie, C. (2017). Learning to ask: Neural question generation for reading comprehension. arXiv preprint arXiv:1705.00106.
Golub, D., Huang, P.-S., He, X., & Deng, L. (2017). Two-stage synthesis networks for transfer learning in machine comprehension. arXiv preprint arXiv:1706.09789.
Heilman, M., & Smith, N. A. (2009). Question generation via overgenerating transformations and ranking. Retrieved from
Heilman, M., & Smith, N. A. (2010). Good Question! Statistical Ranking for Question Generation. Paper presented at the the Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Los Angeles, California.
Howard, J., & Ruder, S. (2018). Universal language model fine-tuning for text classification. arXiv preprint arXiv:1801.06146.
Hrastinski, S. (2008). Asynchronous and synchronous e-learning. Educause quarterly, 31(4), 51-55.
Huang, Y.-T., Tseng, Y.-M., Sun, Y. S., & Chen, M. C. (2014). TEDQuiz: automatic quiz generation for TED talks video clips to assess listening comprehension. Paper presented at the 2014 IEEE 14Th international conference on advanced learning technologies. Athens, Greece
Kaplan, A. M., & Haenlein, M. (2016). Higher education and the digital revolution: About MOOCs, SPOCs, social media, and the Cookie Monster. Business Horizons, 59(4), 441-450.
Kedia, A., Chinthakindi, S. C., Back, S., Lee, H., & Choo, J. (2019). ASGen: Answer-containing Sentence Generation to Pre-Train Question Generator for Scale-up Data in Question Answering. In International conference on learning representations. Addis Ababa, Ethiopia
Kumar, G., Banchs, R., & D’Haro, L. F. (2015). Revup: Automatic Gap-Fill Question Generation from Educational Texts. Paper presented at the the Tenth Workshop on Innovative Use of NLP for Building Educational Applications, Denver, Colorado.
Kumar, V., Ramakrishnan, G., & Li, Y. F. (2018). A Framework for Automatic Question Generation from Text Using Deep Reinforcement Learning. arXiv preprint arXiv:1808.04961.
Lai, G., Xie, Q., Liu, H., Yang, Y., & Hovy, E. (2017). Race: Large-scale reading comprehension dataset from examinations. arXiv preprint arXiv:1704.04683.
Lauren, P., Qu, G., Yang, J., Watta, P., Huang, G.-B., & Lendasse, A. (2018). Generating Word Embeddings from an Extreme Learning Machine for Sentiment Analysis and Sequence Labeling Tasks. Cognitive Computation, 10(4), 625-638.
Levy, O., Goldberg, Y., & Dagan, I. (2015). Improving Distributional Similarity with Lessons Learned from Word Embeddings. Transactions of the Association for Computational Linguistics, 3, 211-225.
Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., . . . Zettlemoyer, L. (2019). Bart: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. arXiv preprint arXiv:1910.13461.
Lin, K.-Y. (2015). Evaluating the effect of a clicker in an information literacy course for college nursing students in Taiwan. CIN: Computers, Informatics, Nursing, 33(3), 115-121.
Lindberg, D., Popowich, F., Nesbit, J., & Winne, P. (2013). Generating Natural Language Questions to Support Learning On-Line. Paper presented at the the 14th European Workshop on Natural Language Generation, Sofia, Bulgaria.
Lindberg, D., Popowich, F., Nesbit, J., & Winne, P. (2013). Generating natural language questions to support learning on-line. Paper presented at the Proceedings of the 14th European Workshop on Natural Language Generation. Sofia, Bulgaria
Liu, M., Calvo, R. A., Aditomo, A., & Pizzato, L. A. (2012). Using Wikipedia and Conceptual Graph Structures to Generate Questions for Academic Writing Support. IEEE Transactions on Learning Technologies, 5(3), 251-263.
Liu, M., Rus, V., & Liu, L. (2017). Automatic Chinese Multiple Choice Question Generation Using Mixed Similarity Strategy. IEEE Transactions on Learning Technologies, 11(2), 193-202.
Liu, S., Li, Z., Zhang, Y., & Cheng, X. (2019). Introduction of key problems in long-distance learning and training. Mobile Networks and Applications, 24(1), 1-4.
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013). Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781.
Mitkov, R. (2003). Computer-aided generation of multiple-choice tests. Paper presented at the Proceedings of the HLT-NAACL 03 workshop on Building educational applications using natural language processing.USA.
Mostow, J., & Jang, H. (2012). Generating diagnostic multiple choice comprehension cloze questions. Paper presented at the Proceedings of the Seventh Workshop on Building Educational Applications Using NLP. Montréal, Canada
Naili, M., Chaibi, A. H., & Ghezala, H. H. B. (2017). Comparative Study of Word Embedding Methods in Topic Segmentation. Procedia Computer Science, 112, 340-349.
Nema, P., & Khapra, M. M. (2018). Towards a Better Metric for Evaluating Question Generation Systems. Paper presented at the the 2018 Conference on Empirical Methods in Natural Language Processing. Brussels, Belgium
Pennington, J., Socher, R., & Manning, C. (2014). Glove: Global Vectors for Word Representation. Paper presented at the Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP). Doha, Qatar.
Perveen, A. (2016). Synchronous and asynchronous e-language learning: A case study of virtual university of Pakistan. Open Praxis, 8(1), 21-39.
Peters, M. E., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K., & Zettlemoyer, L. (2018). Deep contextualized word representations. arXiv preprint arXiv:1802.05365.
Petzold, A. M. (2020). Letter to the Editor: Resources and recommendations for a quick transition to online instruction in physiology. Advances in physiology education, 44(2), 217-219.
Radford, A., Narasimhan, K., Salimans, T., & Sutskever, I. (2018). Improving language understanding with unsupervised learning.
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., . . . Liu, P. J. (2019). Exploring the limits of transfer learning with a unified text-to-text transformer. arXiv preprint arXiv:1910.10683.
Rajpurkar, P., Zhang, J., Lopyrev, K., & Liang, P. (2016). Squad: 100,000+ Questions for Machine Comprehension of Text. Paper presented at the the Empirical Methods in Natural Language Processing (EMNLP), Austin, Texas, USA.
Reimers, N., & Gurevych, I. (2019). Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv preprint arXiv:1908.10084.
Roh, H.-K., & Lee, K.-H. (2017). A Basic Performance Evaluation of the Speech Recognition APP of Standard Language and Dialect using Goolgle, Naver, and Daum KAKAO APIs. Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology, 7(12), 819-829.
Schnabel, T., Labutov, I., Mimno, D., & Joachims, T. (2015). Evaluation methods for unsupervised word embeddings. Paper presented at the Proceedings of the 2015 conference on empirical methods in natural language processing. Lisbon, Portugal
Spector, J. M., Merrill, M. D., Elen, J., & Bishop, M. J. (2014). Handbook of research on educational communications and technology: Springer.
Sutskever, I., Vinyals, O., & Le, Q. V. (2014). Sequence to Sequence Learning with Neural Networks. Paper presented at the the Advances in neural information processing systems, Montreal, Quebec, Canada.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., . . . Polosukhin, I. (2017). Attention is all you need. Paper presented at the Advances in neural information processing systems.
Wang, W., Hao, T., & Liu, W. (2007). Automatic question generation for learning evaluation in medicine. Paper presented at the International conference on web-based learning. United Kingdom
Wang, Z., Lan, A. S., Nie, W., Waters, A. E., Grimaldi, P. J., & Baraniuk, R. G. (2018). QG-Net: a Data-Driven Question Generation Model for Educational Content. Paper presented at the the Fifth Annual ACM Conference on Learning at Scale, London, United Kingdom.
Xu, K., Zhang, M., Li, J., Du, S. S., Kawarabayashi, K.-i., & Jegelka, S. (2020). How neural networks extrapolate: From feedforward to graph neural networks. arXiv preprint arXiv:2009.11848.
Yepes, A. J. (2017). Word embeddings and recurrent neural networks based on Long-Short Term Memory nodes in supervised biomedical word sense disambiguation. Journal of biomedical informatics, 73, 137-147.
Young, T., Hazarika, D., Poria, S., & Cambria, E. (2018). Recent Trends in Deep Learning Based Natural Language Processing. ieee Computational intelligenCe magazine, 13(3), 55-75.
Zhou, Q., Yang, N., Wei, F., Tan, C., Bao, H., & Zhou, M. (2017). Neural question generation from text: A preliminary study. Paper presented at the National CCF Conference on Natural Language Processing and Chinese Computing.Dalian,China.