Advertisement

Encoder-Decoder Network with Cross-Match Mechanism for Answer Selection

  • Zhengwen Xie
  • Xiao Yuan
  • Jiawei Wang
  • Shenggen JuEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11856)

Abstract

Answer selection (AS) is an important subtask of question answering (QA) that aims to choose the most suitable answer from a list of candidate answers. Existing AS models usually explored the single-scale sentence matching, whereas a sentence might contain semantic information at different scales, e.g. Word-level, Phrase-level, or the whole sentence. In addition, these models typically use fixed-size feature vectors to represent questions and answers, which may cause information loss when questions or answers are too long. To address these issues, we propose an Encoder-Decoder Network with Cross-Match Mechanism (EDCMN) where questions and answers that represented by feature vectors with fixed-size and dynamic-size are applied for multiple-perspective matching. In this model, Encoder layer is based on the “Siamese” network and Decoder layer is based on the “matching-aggregation” network. We evaluate our model on two tasks: Answer Selection and Textual Entailment. Experimental results show the effectiveness of our model, which achieves the state-of-the-art performance on WikiQA dataset.

Keywords

Answer selection Multi-Perspective Cross-Match Mechanism 

Notes

Acknowledgements

This research was partially supported by the Sichuan Science and Technology Program under Grant Nos. 2018GZ0182, 2018GZ0093 and 2018GZDZX0039.

References

  1. 1.
    Bachrach, Y., et al.: An attention mechanism for neural answer selection using a combined global and local view. In: Proceedings of International Conference on Tools with Artificial Intelligence, ICTAI, November 2017, pp. 425–432 (2018).  https://doi.org/10.1109/ICTAI.2017.00072
  2. 2.
    Hochreiter, S.: Recurrent cascade-correlation and neural sequence chunking. LSTM, vol. 9, no. 8, pp. 1–32 (1997)Google Scholar
  3. 3.
    Chen, Q., et al.: Enhanced LSTM for Natural Language Inference, pp. 1657–1668 (2016)Google Scholar
  4. 4.
    Feng, M., et al.: Applying deep learning to answer selection: a study and an open task. In: 2015 Proceedings of IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2015, pp. 813–820 (2016).  https://doi.org/10.1109/ASRU.2015.7404872
  5. 5.
    He, H., Lin, J.: Pairwise word interaction modeling with deep neural networks for semantic similarity measurement, pp. 937–948 (2016).  https://doi.org/10.18653/v1/n16-1108
  6. 6.
    Heilman, M., Smith, N.: Tree edit models for recognizing textual entailments, paraphrases, and answers to questions. In: Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, pp. 1011–1019, June 2010Google Scholar
  7. 7.
    Khot, T., et al.: SCITAIL: a textual entailment dataset from science question answering. In: AAAI Conference on Artificial Intelligence (2018)Google Scholar
  8. 8.
    Lin, W.S., et al.: Utilizing different word representation methods for twitter data in adverse drug reactions extraction. In: 2015 Conference on Technologies and Applications of Artificial Intelligence, TAAI 2015, pp. 260–265 (2016).  https://doi.org/10.1109/TAAI.2015.7407070
  9. 9.
    Lin, Z., et al.: A structured self-attentive sentence embedding, pp. 1–15 (2017)Google Scholar
  10. 10.
    Yang, Y., Yih, W.T., Meek, C.: WIKI QA : a challenge dataset for open-domain question answering, September 2015, pp. 2013–2018 (2018)Google Scholar
  11. 11.
    Rücklé, A., Gurevych, I.: Representation learning for answer selection with LSTM-based importance weighting. In: Proceedings of 12th International Conference on Computational Semantics (IWCS 2017) (2017, to appear)Google Scholar
  12. 12.
    dos Santos, C., et al.: Attentive pooling networks. Cv (2016)Google Scholar
  13. 13.
    Severyn, A.: Rank with CNN (2014).  https://doi.org/10.1145/2766462.2767738
  14. 14.
    Shen, G., et al.: Inter-weighted alignment network for sentence pair modeling, pp. 1179–1189 (2018).  https://doi.org/10.18653/v1/d17-1122
  15. 15.
    Shen, T., et al.: DiSAN: directional self-attention network for RNN/CNN-free language understanding (2017)Google Scholar
  16. 16.
    Shen, Y., et al.: A latent semantic model with convolutional-pooling structure for information retrieval. In: Proceedings of 23rd ACM International Conference on Conference on Information and Knowledge Management - CIKM 2014, pp. 101–110 (2014).  https://doi.org/10.1145/2661829.2661935
  17. 17.
    Tan, M., et al.: LSTM-based deep learning models for non-factoid answer selection, vol. 1, pp. 1–11 (2015)Google Scholar
  18. 18.
    Tay, Y., et al.: Multi-cast attention networks for retrieval-based question answering and response prediction (2018)Google Scholar
  19. 19.
    Tran, N.K., Niedereée, C.: Multihop attention networks for question answer matching, pp. 325–334 (2018).  https://doi.org/10.1145/3209978.3210009
  20. 20.
    Wang, B., et al.: Inner attention based recurrent neural networks for answer selection, pp. 1288–1297 (2016).  https://doi.org/10.18653/v1/p16-1122
  21. 21.
    Wang, M., et al.: What is the jeopardy model? A quasi-synchronous grammar for QA. In: Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pp. 22–32, June 2007Google Scholar
  22. 22.
    Wang, S., Jiang, J.: A compare-aggregate model for matching text sequences, pp. 1–11 (2016)Google Scholar
  23. 23.
    Wang, Z., et al.: Bilateral multi-perspective matching for natural language sentences. In: International Joint Conference on Artificial Intelligence, pp. 4144–4150 (2017)Google Scholar
  24. 24.
    Wang, Z., et al.: Sentence similarity learning by lexical decomposition and composition. Challenge 2 (2016)Google Scholar
  25. 25.
    Wang, Z., Ittycheriah, A.: FAQ-based question answering via word alignment, p. 1 (2015)Google Scholar
  26. 26.
    Yang, L., et al.: aNMM: ranking short answer texts with attention-based neural matching model (2018)Google Scholar
  27. 27.
    Yin, W., et al.: ABCNN: attention-based convolutional neural network for modeling sentence pairs (2015)Google Scholar
  28. 28.
    He, K., et al.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778, December 2016.  https://doi.org/10.1109/CVPR.2016.90

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Zhengwen Xie
    • 1
  • Xiao Yuan
    • 1
  • Jiawei Wang
    • 1
  • Shenggen Ju
    • 1
    Email author
  1. 1.College of Computer ScienceSichuan UniversityChengduChina

Personalised recommendations