SpringerLink
Log in

Learning Enhancement Using Question-Answer Generation for e-Book Using Contrastive Fine-Tuned T5

  • Conference paper
  • First Online:
Big Data Analytics (BDA 2022)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 13773))

Included in the following conference series:

Abstract

The rise of E-learning systems offers a vast amount of free educational content for every inquisitive e-learner. However, reading only e-content does not makes their learning effective. Posing appropriate questions during the reading process can aid in the learner’s comprehension. We present a novel approach to create educational Question-Answers on eBook content. The model first summarizes key information from an input document, which is then used for creating relevant Question-Answers (QAs). We build our educational Question-Answers generation model by fine-tuning a pretrained LM (T5) using maximum likelihood estimation. We also present a contrastive fine-tuning method, in which the contrastive loss is added between the positive and negative training feature pairs during the fine-tuning process. The extensive evaluation methods on QA dataset (FairytaleQA and HotPotQA) and NCERT e-book, demonstrate the effectiveness and practicability of our eQA model.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
EUR 32.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or Ebook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
EUR 29.95
Price includes VAT (France)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
EUR 53.49
Price includes VAT (France)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
EUR 68.56
Price includes VAT (France)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free ship** worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Notes

  1. 1.

    https://ncert.nic.in/textbook.php.

  2. 2.

    https://huggingface.co/docs/transformers/model_doc/t5.

  3. 3.

    https://ncert.nic.in/textbook.php.

  4. 4.

    https://github.com/Tiiiger/bert_score.

  5. 5.

    https://github.com/google-research/google-research/tree/master/rouge.

References

  1. Banerjee, S., Lavie, A.: METEOR: an automatic metric for MT evaluation with improved correlation with human judgments. In: Proceedings of the ACL Workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization, Ann Arbor, Michigan, June 2005, pp. 65–72. Association for Computational Linguistics (2005)

    Google Scholar 

  2. Bishop, C.M.: Pattern Recognition and Machine Learning (Information Science and Statistics). Springer, Heidelberg (2006)

    MATH  Google Scholar 

  3. Chan, Y.-H., Fan, Y.-C.: A recurrent BERT-based model for question generation. In: Proceedings of the 2nd Workshop on Machine Reading for Question Answering, Hong Kong, China, November 2019, pp. 154–162. Association for Computational Linguistics (2019)

    Google Scholar 

  4. Chen, G., Yang, J., Gasevic, D.: A comparative study on question-worthy sentence selection strategies for educational question generation. In: Isotani, S., Millán, E., Ogan, A., Hastings, P., McLaren, B., Luckin, R. (eds.) AIED 2019. LNCS (LNAI), vol. 11625, pp. 59–70. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-23204-7_6

    Chapter  Google Scholar 

  5. Cho, W.S., et al.: Contrastive multi-document QG. In: Proceedings of the 16th European Conference, Chapter of the ACL, pp. 12–30 (2021)

    Google Scholar 

  6. De Kuthy, K., Kannan, M., Ponnusamy, H.S., Meurers, D.: Towards automatically generating questions under discussion to link information and discourse structure. In: Proceedings of the 28th International Conference on Computational Linguistics, pp. 5786–5798 (2020)

    Google Scholar 

  7. Demszky, D., Guu, K., Liang, P.: Transforming question answering datasets into natural language inference datasets. CoRR, abs/1809.02922 (2018)

    Google Scholar 

  8. Devlin, J., Chang, M.-W., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding. In: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, (Long and Short Papers), Minneapolis, Minnesota, June 2019, vol. 1, pp. 4171–4186. Association for Computational Linguistics (2019)

    Google Scholar 

  9. Dhole, K.D., Manning, C.D.: Syn-QG: syntactic and shallow semantic rules for question generation. ar**v, abs/2004.08694 (2020)

    Google Scholar 

  10. Dong, L., et al.: Unified Language Model Pre-training for Natural Language Understanding and Generation. Curran Associates Inc., Red Hook (2019)

    Google Scholar 

  11. Du, X., Shao, J., Cardie, C.: Learning to ask: neural question generation for reading comprehension. CoRR, abs/1705.00106 (2017)

    Google Scholar 

  12. Ebersbach, M., Feierabend, M., Nazari, K.B.B.: Comparing the effects of generating questions, testing, and restudying on students’ long-term recall in university learning. Appl. Cogn. Psychol. 34(3), 724–736 (2020)

    Article  Google Scholar 

  13. Kurdi, G., Leo, J., Parsia, B., Sattler, U., Al-Emari, S.: A systematic review of automatic question generation for educational purposes. Int. J. Artif. Intell. Educ. 30, 121–204 (2020)

    Article  Google Scholar 

  14. Golinkoff, R.M.: Language matters: denying the existence of the 30-million-word gap has serious consequences. Child Dev. 90(3), 985–992 (2019)

    Article  Google Scholar 

  15. Gretz, S., Bilu, Y., Cohen-Karlik, E., Slonim, N.: The workweek is the best time to start a family - a study of GPT-2 based claim generation. In: Findings of the Association for Computational Linguistics, EMNLP 2020, November 2020. Association for Computational Linguistics, pp. 528–544 (2020)

    Google Scholar 

  16. Gwet, K.L.: On the Krippendorff’s alpha coefficient (2011)

    Google Scholar 

  17. Ganotice, F.A., Jr., Downing, K., Mak, T., Chan, B., Lee, W.Y.: Enhancing parent-child relationship through dialogic reading. Educ. Stud. 43(1), 51–66 (2017)

    Article  Google Scholar 

  18. Kočiský, T., et al.: The NarrativeQA reading comprehension challenge. Trans. Assoc. Comput. Linguist. 6, 317–328 (2018)

    Article  Google Scholar 

  19. Krishna, K., Iyyer, M.: Generating question-answer hierarchies. In: Proceedings of the 57th Annual Meeting of the ACL, Florence, Italy, pp. 2321–2334, July 2019. Association for Computational Linguistics (2019)

    Google Scholar 

  20. Kuamr, S., Chauhan, A.: Augmenting textbooks with CQA question-answers and annotated YouTube videos to increase its relevance. Neural Process Lett. (2022)

    Google Scholar 

  21. Kumar, S., Chauhan, A.: Enriching textbooks by question-answers using CQA. In: 2019 IEEE Region 10 Conference (TENCON), TENCON 2019, pp. 707–714 (2019)

    Google Scholar 

  22. Kumar, S., Chauhan, A.: Recommending question-answers for enriching textbooks. In: Bellatreche, L., Goyal, V., Fujita, H., Mondal, A., Reddy, P.K. (eds.) BDA 2020. LNCS, vol. 12581, pp. 308–328. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-66665-1_20

    Chapter  Google Scholar 

  23. Kumar, S., Chauhan, A.: A finetuned language model for recommending cQA-QAs for enriching textbooks. In: Karlapalem, K., et al. (eds.) PAKDD 2021. LNCS (LNAI), vol. 12713, pp. 423–435. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-75765-6_34

    Chapter  Google Scholar 

  24. Labutov, I., Basu, S., Vanderwende, L.: Deep questions without deep understanding. In: Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), Bei**g, China, July 2015, pp. 889–898. Association for Computational Linguistics (2015)

    Google Scholar 

  25. Lewis, P., et al.: PAQ: 65 million probably-asked questions and what you can do with them. Trans. Assoc. Comput. Linguist. 9, 1098–1115 (2021)

    Article  Google Scholar 

  26. Li, M., et al.: Timeline summarization based on event graph compression via time-aware optimal transport. In: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, Punta Cana, Dominican Republic, pp. 6443–6456, November 2021. Association for Computational Linguistics (2021)

    Google Scholar 

  27. Lin, C.-Y.: ROUGE: a package for automatic evaluation of summaries. In: Text Summarization Branches Out, Barcelona, Spain, July 2004, pp. 74–81. Association for Computational Linguistics (2004)

    Google Scholar 

  28. Lin, T.-Y., Goyal, P., Girshick, R., He, K., DollÁr, P.: Focal loss for dense object detection. IEEE Trans. Pattern Anal. Mach. Intell. 42(2), 318–327 (2020)

    Article  Google Scholar 

  29. Lyu, C., Shang, L., Graham, Y., Foster, J., Jiang, X., Liu, Q.: Improving unsupervised question answering via summarization-informed question generation. In: Proceedings of the EMNLP, pp. 4134–4148 (2021)

    Google Scholar 

  30. Nguyen, T., et al.: MS MARCO: a human generated machine reading comprehension dataset. CoRR, abs/1611.09268 (2016)

    Google Scholar 

  31. Pan, L., **e, Y., Feng, Y., Chua, T.-S., Kan, M.-Y.: Semantic graphs for generating deep questions. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, July 2020, pp. 1463–1475. Association for Computational Linguistics (2020)

    Google Scholar 

  32. Pan, L., **e, Y., Feng, Y., Chua, T.-S., Kan, M.-Y.: Semantic graphs for generating deep questions. In: Proceedings of the 58th Annual Meeting of the ACL, pp. 1463–1475 (2020)

    Google Scholar 

  33. Papineni, K., Roukos, S., Ward, T., Zhu, W.-J.: Bleu: a method for automatic evaluation of machine translation. In: Proceedings of the 40th Annual Meeting on Association for Computational Linguistics, ACL 2002, USA, pp. 311–318. Association for Computational Linguistics (2002)

    Google Scholar 

  34. Pyatkin, V., Roit, P., Michael, J., Tsarfaty, R., Goldberg, Y., Dagan, I.: Asking it all: generating contextualized questions for any semantic role. CoRR, abs/2109.04832 (2021)

    Google Scholar 

  35. Qi, W., et al.: ProphetNet: predicting future n-gram for sequence-to-sequence pre-training. In: EMNLP 2020, pp. 2401–2410 (2020)

    Google Scholar 

  36. Radford, A., et al.: Learning transferable visual models from natural language supervision. CoRR, abs/2103.00020 (2021)

    Google Scholar 

  37. Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., Sutskever, I.: Language Models are Unsupervised Multitask Learners (2019)

    Google Scholar 

  38. Ruan, S., et al.: Quizbot: a dialogue-based adaptive learning system for factual knowledge. In: Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, CHI 2019, pp. 1–13. Association for Computing Machinery, New York (2019)

    Google Scholar 

  39. Rush, A.M., Chopra, S., Weston, J.: A neural attention model for abstractive sentence summarization. In: Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, Lisbon, Portugal, September 2015, pp. 379–389. Association for Computational Linguistics (2015)

    Google Scholar 

  40. Scialom, T., Piwowarski, B., Staiano, J.: Self-attention architectures for answer-agnostic neural question generation. In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy, July 2019, pp. 6027–6032. Association for Computational Linguistics (2019)

    Google Scholar 

  41. Stasaski, K., Hearst, M.A.: Multiple choice question generation utilizing an ontology. In: Proceedings of the 12th Workshop on Innovative Use of NLP for Building Educational Applications, Copenhagen, Denmark, September 2017, pp. 303–312. Association for Computational Linguistics (2017)

    Google Scholar 

  42. Tuan, L.A., Shah, D.J., Barzilay, R.: Capturing greater context for question generation. CoRR, abs/1910.10274 (2019)

    Google Scholar 

  43. Wang, D., Liu, P., Zheng, Y., Qiu, X., Huang, X.: Heterogeneous graph neural networks for extractive document summarization. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, July 2020, pp. 6209–6219. Association for Computational Linguistics (2020)

    Google Scholar 

  44. Wang, T., Yuan, X., Trischler, A.: A joint model for question answering and question generation. In: Learning to Generate Natural Language Workshop, ICML 2017 (2017)

    Google Scholar 

  45. Wang, Z., Lan, A.S., Nie, W., Waters, A.E., Grimaldi, P.J., Baraniuk, R.G.: QG-Net: a data-driven question generation model for educational content. In: Proceedings of the Fifth Annual ACM Conference on Learning at Scale, L@S 2018. Association for Computing Machinery, New York (2018)

    Google Scholar 

  46. Xu, J., Gan, Z., Cheng, Y., Liu, J.: Discourse-aware neural extractive text summarization. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp. 5021–5031, July 2020. Association for Computational Linguistics (2020)

    Google Scholar 

  47. Xu, Y., Wang, D., Collins, P., Lee, H., Warschauer, M.: Same benefits, different communication patterns: comparing children’s reading with a conversational agent vs. a human partner. Comput. Educ. (2021)

    Google Scholar 

  48. Xu, Y., et al.: Fantastic questions and where to find them: FairytaleQA - an authentic dataset for narrative comprehension. Association for Computational Linguistics (2022)

    Google Scholar 

  49. Yang, Z., et al.: HotpotQA: a dataset for diverse, explainable multi-hop question answering. CoRR, abs/1809.09600 (2018)

    Google Scholar 

  50. Zhang, T., Kishore, V., Wu, F., Weinberger, K.Q., Artzi, Y.: BERTscore: evaluating text generation with BERT. CoRR, abs/1904.09675 (2019)

    Google Scholar 

  51. Zhao, Y., Ni, X., Ding, Y., Ke, Q.: Paragraph-level neural question generation with maxout pointer and gated self-attention networks. In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium, October–November 2018, pp. 3901–3910. Association for Computational Linguistics (2018)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Indian Institute of Information Technology Dharwad, Dharwad, Karnataka, India

    Shobhan Kumar, Arun Chauhan & Pavan Kumar C.

Authors
  1. Shobhan Kumar
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Arun Chauhan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Pavan Kumar C.
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Shobhan Kumar .

Editor information

Editors and Affiliations

  1. Indian Institute of Technology-Roorkee, Roorkee, India

    Partha Pratim Roy

  2. IBM Research, Gurugram, India

    Arvind Agarwal

  3. Southwest Jiaotong University, Chengdu, China

    Tianrui Li

  4. International Institute of Information Technology - Hyderabad, Hyderabad, India

    P. Krishna Reddy

  5. The University of Aizu, Fukushima, Japan

    R. Uday Kiran

Appendices

Appendix

A BERT Score for Semantic Match

Besides the standard evaluation metrics such as, METEOR and Rouge-L scores, we also employ BERTScore [50] to assess the semantic similarity of questions generated by our QG model with the ground-truth questions, and present the average precision, recall, and F1 values. Table 12 outlines the computed BERT scores for top k generated questions.The results show that our model can generate relevant questions for the given text.

Table 12. The comparison results (BERT score) of our model when k = 1, 3, 5 (top k questions) on FairytaleQA test data.
Full size table

B Few More Examples of Generated QAs for Text Document

Two more samples (input text), the ground truth QAs, generated QAs by CBQA model, and our edu-QA generation model. Tables 13 depict the sample text from NCERT CCT eBook, generated summary, QAs by our model and generated QAs by CBQA model [25]. Table 14 contain a generated QAs for the sample input text from FairytaleQA dataset, associated ground truth QAs, and generated summary, QAs by our model and generated QAs by CBQA model [25].

Table 13. The generated QAs for the sample input text from NCERT e-BOOK (Subject: CCT- Chap. 13) data. Here, the results of PAQ baseline model, our edu-QA generating mechanism, and the ground truth QAs are shown.
Full size table
Table 14. The generated QAs for the sample input text from FairytaleQA [48] data. Here, the results of CBQA baseline model (Patrick et al., 2021) [25], our edu-QA generating mechanism, and the ground truth QAs are shown.
Full size table

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Kumar, S., Chauhan, A., Kumar C., P. (2022). Learning Enhancement Using Question-Answer Generation for e-Book Using Contrastive Fine-Tuned T5. In: Roy, P.P., Agarwal, A., Li, T., Krishna Reddy, P., Uday Kiran, R. (eds) Big Data Analytics. BDA 2022. Lecture Notes in Computer Science, vol 13773. Springer, Cham. https://doi.org/10.1007/978-3-031-24094-2_5

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-24094-2_5

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-24093-5

  • Online ISBN: 978-3-031-24094-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation