Advertisement

A Two-Step Neural Network Approach to Passage Retrieval for Open Domain Question Answering

  • Andrés Rosso-Mateus
  • Fabio A. González
  • Manuel Montes-y-Gómez
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10657)

Abstract

Passage retrieval is an important subtask of question answering. Given a question and a set of candidate passages, the goal is to rank them according to their relevance to the question. This work presents a two-stage approach for solving this problem. Both stages are based on convolutional neural network architecture with a reduced set of parameters. In the first stage the network is used to identify the degree of similarity between question and candidate answers, then, in the second stage, the result of the first stage is used to re-rank the answers according to their similarity with the initial best-ranked answer in such a way that the most similar candidate answers are moved up. This approach is analogous to a pseudo-relevance feedback strategy. The experimental results suggest that the proposed method is competitive with the state-of-the-art methods, achieving a remarkable performance in three evaluation datasets.

Keywords

Question answering Passage retrieval Answer ranking Pseudo-relevance feedback 

References

  1. 1.
    ACL, Association for Computational Linguistics: Question Answering (State of the art) (2007). http://www.aclweb.org/aclwiki/index.php. Accessed 1 May 2017
  2. 2.
    Bird, S.: NLTK: the natural language toolkit. In: Proceedings of the COLING/ACL on Interactive Presentation Sessions, vol. 1, pp. 69–72. ACL (2006)Google Scholar
  3. 3.
    Dong, L., Wei, F., Zhou, M., Xu, K.: Question answering over freebase with multi-column convolutional neural networks. In: ACL, vol. 1, pp. 260–269 (2015)Google Scholar
  4. 4.
    Etzioni, O.: Search needs a shake-up. Nature 476(7358), 25–26 (2011)CrossRefGoogle Scholar
  5. 5.
    He, H., Lin, J.J.: Pairwise word interaction modeling with deep neural networks for semantic similarity measurement. In: HLT-NAACL, vol. 1, pp. 937–948 (2016)Google Scholar
  6. 6.
    Le, Q.V., Mikolov, T.: Distributed representations of sentences and documents. In: ICML, vol. 14, pp. 1188–1196 (2014)Google Scholar
  7. 7.
    Levenshtein, V.I.: Binary codes capable of correcting deletions, insertions, and reversals. In: Soviet Physics Doklady, vol. 10, pp. 707–710 (1966)Google Scholar
  8. 8.
    Liu, F., Pennell, D., Liu, F., Liu, Y.: Unsupervised approaches for automatic keyword extraction using meeting transcripts. In: Proceedings of Human Language Technologies, vol. 1, pp. 620–628. ACL (2009)Google Scholar
  9. 9.
    Mikolov, T., Chen, K., Corrado, G.S., Dean, J.: Efficient estimation of word representations in vector space (2013)Google Scholar
  10. 10.
    Riezler, S., Vasserman, A., Tsochantaridis, I., Mittal, V., Liu, Y.: Statistical machine translation for query expansion in answer retrieval. In: ACL (2007)Google Scholar
  11. 11.
    Severyn, A., Moschitti, A.: Learning to rank short text pairs with convolutional deep neural networks. In: 38th ACM SIGIR (2015)Google Scholar
  12. 12.
    Wang, M., Smith, N.A., Mitamura, T.: What is the jeopardy model? A quasi-synchronous grammar for QA, vol. 7, pp. 22–32 (2007)Google Scholar
  13. 13.
    Wu, Z., Palmer, M.: Verbs semantics and lexical selection. In: 32nd Proceedings ACL, vol. 1, pp. 133–138. Association for Computational Linguistics (1994)Google Scholar
  14. 14.
    Yang, L., Ai, Q., et al.: aNMM: ranking short answer texts with attention-based neural matching model. In: Proceedings of the 25th ACM ICIKM. ACM (2016)Google Scholar
  15. 15.
    Yang, Y., Yih, W., Meek, C.: WikiQA: a challenge dataset for open-domain question answering. In: EMNLP, vol. 1, pp. 2013–2018 (2015)Google Scholar
  16. 16.
    Yu, L., Hermann, K.M., Blunsom, P., Pulman, S.: Deep learning for answer sentence selection. In: NIPS Deep Learning Workshop (2014)Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.MindLab Research GroupUniversidad Nacional de ColombiaBogotáColombia
  2. 2.Computer Science DepartmentInstituto Nacional de Astrofísica, Óptica y ElectrónicaPueblaMexico

Personalised recommendations