Skip to main content

Advertisement

Springer Nature Link
Log in

Stacked dynamic memory-coattention network for answering why-questions in Arabic

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

Answering non-factoid questions, especially why-questions, poses a challenge for traditional question answering systems (QASs) that are predominantly designed for fact-based queries. Recent advancements in QASs have incorporated attention and memory mechanisms to better capture the intricate relationship between context and query, resulting in more precise answers. Two prominent mechanisms in this domain are the dynamic memory network (DMN) and the dynamic coattention network (DCN). In this study, we propose a novel model named stacked dynamic memory-coattention network (SDMCN) that combines the memory mechanism of DMN with the coattention mechanism of DCN to extract answers for why-questions in Arabic. Our model was evaluated across three distinct Arabic why-question datasets: DAWQAS, LEMAZA, and WA. It achieved accuracies of 81.85%, 83.6%, and 89%, and \(F\)-scores of 80.55%, 82.42%, and 87.52% in the DAWQAS, LEMAZA, and WA datasets, respectively. These results not only underscore the SDMCN’s effectiveness but also highlight its superiority over the individual baseline models of DMN and DCN, with the SDMCN achieving a mean \(F\)-score of 83.50% and a weighted \(F\)-score of 82.36%. These results demonstrate the effectiveness of our model for answering why-questions in Arabic.

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

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Algorithm 1
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Data availability

The WA dataset described in this work is available at https://github.com/talwaneen/WA-dataset.git.

Notes

  1. DAWQAS is freely available at http://github.com/masum/DAWQAS.

  2. WA can be downloaded from https://github.com/talwaneen/WA-dataset.

References

  1. Ray SK, Shaalan K (2016) A review and future perspectives of Arabic question answering systems. IEEE Trans Knowl Data Eng 28(12):3169–3190

    Article  Google Scholar 

  2. Alwaneen TH, Azmi AM, Aboalsamh HA, Cambria E, Hussain A (2021) Arabic question answering system: a survey. Artif Intell Rev 55(1):207–253

    Article  Google Scholar 

  3. Xiong C, Zhong V, Socher R (2016) Dynamic coattention networks for question answering. In: International conference on learning representations (ICLR)

  4. Xiong C, Merity S, Socher R. (2016) Dynamic memory networks for visual and textual question answering. In: International conference on machine learning, pp 2397–2406

  5. Kumar A, Irsoy O, Ondruska P, Iyyer M, Bradbury J, Gulrajani I et al (2016) Ask me anything: dynamic memory networks for natural language processing. In: International conference on machine learning. PMLR, pp 1378–1387

  6. Santana A, Colombini E (2021) Neural attention models in deep learning: survey and taxonomy. arXiv preprint arXiv:2112.05909

  7. Azmi AM, Alshenaifi NA (2016) Answering Arabic why-questions: baseline vs. RST-based approach. ACM Trans Inf Syst 35(1):1–19

    Article  Google Scholar 

  8. Mannaa ZM, Azmi AM, Aboalsamh HA (2022) Computer-assisted i‘raab of Arabic sentences for teaching grammar to students. J King Saud Univ Comput Inf Sci 34(10):8909–8926

    Google Scholar 

  9. Samy H, Hassanein EE, Shaalan K (2019) Arabic question answering: a study on challenges, systems, and techniques. Int J Comput Appl 181(44):1–10

    Google Scholar 

  10. Azmi AM, Aljafari EA (2015) Modern information retrieval in Arabic-catering to standard and colloquial Arabic users. J Inf Sci 41(4):506–517

    Article  Google Scholar 

  11. Azmi AM, Aljafari EA (2018) Universal web accessibility and the challenge to integrate informal Arabic users: a case study. Univ Access Inf Soc 17:131–145

    Article  Google Scholar 

  12. Feng M, Xiang B, Glass MR, Wang L, Zhou B (2015) Applying deep learning to answer selection: A study and an open task. In: 2015 IEEE workshop on automatic speech recognition and understanding (ASRU). IEEE, pp 813–820

  13. Cohen D, Croft WB (2016) End to end long short term memory networks for non-factoid question answering. In: Proceedings of the 2016 ACM international conference on the theory of information retrieval, pp 143–146

  14. Tan M, Santos Cd, Xiang B, Zhou B (2015) LSTM-based deep learning models for non-factoid answer selection. arXiv preprint arXiv:1511.04108

  15. Xiong C, Zhong V, Socher R (2017) Dcn+: mixed objective and deep residual coattention for question answering. arXiv preprint arXiv:1711.00106

  16. Kumar A, Mittal T, Manocha D (2020) Mcqa: multimodal co-attention based network for question answering. arXiv preprint arXiv:2004.12238

  17. Rücklé A, Gurevych I (2017) End-to-end non-factoid question answering with an interactive visualization of neural attention weights. In: Proceedings of ACL 2017, system demonstrations, pp 19–24

  18. Nugaliyadde A, Wong KW, Sohel F, Xie H (2017) Reinforced memory network for question answering. In: International conference on neural information processing. Springer, pp 482–490

  19. Kale S, Kulkarni A, Patil R, Haribhakta Y, Bhattacharjee K, Mehta S et al (2018) Open-domain question answering using feature encoded dynamic coattention networks. In: 2018 International conference on advances in computing, communications and informatics (ICACCI). IEEE, pp 1058–1062

  20. Chen Y, Wu L, Zaki MJ (2019) Bidirectional attentive memory networks for question answering over knowledge bases. arXiv preprint arXiv:1903.02188

  21. Wu J, Mu T, Thiyagalingam J, Goulermas JY (2023) Memory-aware attentive control for community question answering with knowledge-based dual refinement. IEEE Trans Syst Man Cybern Syst 53(7):3930–3943

    Article  Google Scholar 

  22. Zhou Z, Kang X, Nishide S, Ren F (2019) IDMN: a two-pass attention mechanism in dynamic memory network. In: 2019 IEEE 6th international conference on cloud computing and intelligence systems (CCIS). IEEE, pp 1–5

  23. Liu M, Zhou F, Chen K, Zhao Y (2021) Co-attention networks based on aspect and context for aspect-level sentiment analysis. Knowl Based Syst 217:106810

    Article  Google Scholar 

  24. Cheng LC, Chen YL, Liao YY (2022) Aspect-based sentiment analysis with component focusing multi-head co-attention networks. Neurocomputing 489:9–17

    Article  Google Scholar 

  25. Biltawi MM, Tedmori S, Awajan A (2021) Arabic question answering systems: gap analysis. IEEE Access 9:63876–63904

    Article  Google Scholar 

  26. Salem Z, Sadek J, Chakkour F, Haskkour N (2010) Automatically finding answers to “Why” and “How to” questions for Arabic language. In: Proceedings of knowledge-based and intelligent information and engineering systems: 14th international conference, KES 2010, Cardiff, UK, September 8–10, 2010. Springer, pp 586–593

  27. Azmi AM, AlShenaifi N (2014) Handling “why” questions in Arabic. In: Proceedings of 2014 fifth international conference on Arabic language processing (CITALA ’14), Morocco, November 26–27

  28. Azmi AM, Alshenaifi NA (2017) LEMAZA: an Arabic why-question answering system. Nat Lang Eng 23(6):877–903

    Article  Google Scholar 

  29. Mann WC, Thompson SA (1988) Rhetorical structure theory: toward a functional theory of text organization. Text 8(3):243–281

    Google Scholar 

  30. Saad MK, Ashour WM (2010) OSAC: Open source Arabic corpora. In: 6th International conference on electrical and computer systems, vol 10. European University of Lefke, Lefke, North Cyprus, pp 25–26

  31. Ahmed W, Anto B (2016) Answer extraction for how and why questions in question answering systems. Int J Comput Eng Res 12(6):18–22

    Google Scholar 

  32. Sadek J, Meziane F (2016) A discourse-based approach for Arabic question answering. ACM Trans Asian Low Resour Lang Inf Process 16(2):1–18

    Article  Google Scholar 

  33. AL-Khawaldeh FT (2019) Answer extraction for why Arabic questions answering systems: EWAQ. World Comput Sci Inf Technol J 5(5):82–86

    Google Scholar 

  34. Mozannar H, Hajal KE, Maamary E, Hajj H (2019) Neural Arabic question answering. In: Proceedings of the 4th Arabic natural language processing workshop. Association for Computational Linguistics. Florence, Italy

  35. Alwaneen T, Azmi A (2023) A deep learning model for answering why-questions in Arabic. Indones J Comput Sci 12(2):401–409

    Article  Google Scholar 

  36. Almuzaini HA, Azmi AM (2020) Impact of stemming and word embedding on deep learning-based Arabic text categorization. IEEE Access 8:127913–127928

    Article  Google Scholar 

  37. Abainia K, Ouamour S, Sayoud H (2017) A novel robust Arabic light stemmer. J Exp Theor Artif Intell 29(3):557–573

    Article  Google Scholar 

  38. Ismail WS, Homsi MN (2018) DAWQAS: a dataset for Arabic why question answering system. Procedia Comput Sci 142:123–131

    Article  Google Scholar 

  39. Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980

  40. Joulin A, Grave E, Bojanowski P, Douze M, Jégou H, Mikolov T (2016) Fasttext. zip: compressing text classification models. arXiv preprint arXiv:1612.03651

Download references

Funding

The authors extend their appreciation to the Deputyship for Research & Innovation, Ministry of Education in Saudi Arabia for funding this research work through the Project No. IFKSUOR3-049-3. The funding sponsors were not involved in any matters related to research.

Author information

Authors and Affiliations

  1. Department of Computer Science, College of Computer and Information Sciences, King Saud University, 11543, Riyadh, Saudi Arabia

    Tahani H. Alwaneen & Aqil M. Azmi

Authors
  1. Tahani H. Alwaneen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Aqil M. Azmi
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Aqil M. Azmi.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by the authors.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Alwaneen, T.H., Azmi, A.M. Stacked dynamic memory-coattention network for answering why-questions in Arabic. Neural Comput & Applic 36, 8867–8883 (2024). https://doi.org/10.1007/s00521-024-09525-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-024-09525-0

Keywords

Search

Navigation

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy