Advertisement

On the Impact of Semantic Roles on Text Comprehension for Question Answering

  • Anca Marginean
  • Gabriela Pricop
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11308)

Abstract

New challenges for question answering are introduced by texts whose understanding require inference and commonsense knowledge. Task 11 - Machine comprehension using Commonsense Knowledge - from SemEval 2018 proposes a corpus of such texts, questions and answers. Since the predicates identified by Semantic Role Labeling aim to capture the semantic of a sentence, they seem appropriate to the task of text comprehension. We propose a Context-Novelty based model for identification of the correct answer for a question. This model relies on the SRL predicates of the text, question and answers and (i) it targets identification of the parts from the text which are relevant to the current question and (ii) it measures how well the answer matches that parts. The performance of the model was evaluated directly by counting the number of correctly answered questions and by its integration to a classical machine learning process.

Keywords

Semantic roles ConceptNet Commonsense reasoning 

References

  1. 1.
    Baker, C.F., Sato, H.: The FrameNet data and software. In: The Companion Volume to the Proceedings of 41st Annual Meeting of the Association for Computational Linguistics (2003)Google Scholar
  2. 2.
    Chen, D., Bolton, J., Manning, C.D.: A thorough examination of the CNN/daily mail reading comprehension task. In: Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 2358–2367. Association for Computational Linguistics (2016)Google Scholar
  3. 3.
    He, L., Lee, K., Lewis, M., Zettlemoyer, L.S.: Deep semantic role labeling: what works and what’s next. In: ACL (2017)Google Scholar
  4. 4.
    MacCartney, B., Manning, C.D.: Natural logic for textual inference. In: Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing, RTE 2007, pp. 193–200 (2007)Google Scholar
  5. 5.
    Merkhofer, E., Henderson, J., Bloom, D., Strickhart, L., Zarrella, G.: MITRE at SemEval-2018 Task 11: commonsense reasoning without commonsense knowledge. In: Proceedings of the 12th International Workshop on Semantic Evaluation, pp. 1078–1082. Association for Computational Linguistics (2018)Google Scholar
  6. 6.
    Ostermann, S., Roth, M., Modi, A., Thater, S., Pinkal, M.: SemEval-2018 Task 11: machine comprehension using commonsense knowledge. In: Proceedings of the 12th International Workshop on Semantic Evaluation, pp. 747–757. Association for Computational Linguistics (2018)Google Scholar
  7. 7.
    Palmer, M., Gildea, D., Kingsbury, P.: The proposition bank: an annotated corpus of semantic roles. Comput. Linguist. 31(1), 71–106 (2005)CrossRefGoogle Scholar
  8. 8.
    Rettinger, A., Schumilin, A., Thoma, S., Ell, B.: Learning a cross-lingual semantic representation of relations expressed in text. In: Gandon, F., Sabou, M., Sack, H., d’Amato, C., Cudré-Mauroux, P., Zimmermann, A. (eds.) ESWC 2015. LNCS, vol. 9088, pp. 337–352. Springer, Cham (2015).  https://doi.org/10.1007/978-3-319-18818-8_21CrossRefGoogle Scholar
  9. 9.
    Roth, M., Lapata, M.: Neural semantic role labeling with dependency path embeddings. In: Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 1192–1202. Association for Computational Linguistics (2016)Google Scholar
  10. 10.
    Walter, S., Unger, C., Cimiano, P.: M-ATOLL: a framework for the lexicalization of ontologies in multiple languages. In: Mika, P., et al. (eds.) ISWC 2014, Part I. LNCS, vol. 8796, pp. 472–486. Springer, Cham (2014).  https://doi.org/10.1007/978-3-319-11964-9_30CrossRefGoogle Scholar
  11. 11.
    Wang, L., Sun, M., Zhao, W., Shen, K., Liu, J.: Yuanfudao at SemEval-2018 Task 11: three-way attention and relational knowledge for commonsense machine comprehension. In: Proceedings of the 12th International Workshop on Semantic Evaluation, pp. 758–762. Association for Computational Linguistics (2018)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Technical University of Cluj-NapocaCluj-NapocaRomania

Personalised recommendations