Propuesta y desarrollo de una aproximación de generación de resúmenes abstractivos multigénero

  1. Esteban García, Alberto
  2. Lloret, Elena
Revue:
Procesamiento del lenguaje natural

ISSN: 1135-5948

Année de publication: 2017

Número: 58

Pages: 53-60

Type: Article

DIALNET GOOGLE SCHOLAR lock_openRUA editor

D'autres publications dans: Procesamiento del lenguaje natural

Résumé

In this paper, the design and development of a multi-genre abstractive summarisation approach is proposed, taking into account information sources belonging to different textual genres. The main objective is to combine the information in all of these genres and produce an abstractive summary, that is, a new coherent text that captures the main ideas about a topic. Specifically, in this research work, information from reviews extracted from TripAdvisor and microblogs gathered via Twitter was used, since the generated summaries were applied to the tourism sector, to provide users with the most and/or less favourable aspects concerning hotels and restaurants. The evaluation carried out with real users shows that the quality of generated summaries meets the standards and therefore, the summaries can be used in real applications.

Références bibliographiques

  • Carenini, G., J. C. K. Cheung, y A. Pauls. 2013. Multi-document summarization of evaluative text. Computational Intelligence, 29(4):545–576.
  • Fernández, J., Y. Gutiérrez, J. M. Gómez, y P. Mart́ınez-Barco. 2015. Social rankings: análisis visual de sentimientos en redes sociales. Procesamiento del Lenguaje Natural, 55:199–202.
  • Gerani, S., Y. Mehdad, G. Carenini, R. T. Ng, y B. Nejat. 2014. Abstractive summarization of product reviews using discourse structure. En Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), páginas 1602–1613, Doha, Qatar, October. Association for Computational Linguistics.
  • Khan, Atif y Salim, N. y. J. K. Y. 2015. A framework for multi-document abstractive summarization based on semantic role labelling. Appl. Soft Comput., 30(C):737– 747, Mayo.
  • Li, W. 2015. Abstractive multi-document summarization with semantic information extraction. En Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, páginas 1908– 1913, Lisbon, Portugal, September. Association for Computational Linguistics.
  • Lloret, E. y E. Boldrini. 2015. Multi-genre summarization: Approach, potentials and challenges. En eChallenges e-2015 Conference, páginas 1–9, Nov.
  • Lloret, Elena y Palomar, M. 2012. Text summarisation in progress: A literature review. Artif. Intell. Rev., 37(1):1–41, Enero.
  • Manning, C. D., M. Surdeanu, J. Bauer, J. Finkel, S. J. Bethard, y D. McClosky. 2014. The Stanford CoreNLP natural language processing toolkit. En Association for Computational Linguistics (ACL) System Demonstrations, páginas 55–60.
  • Saggion, H. 2011. Learning predicate insertion rules for document abstracting. En Computational Linguistics and Intelligent Text Processing 12th International Conference, CICLing 2011, Tokyo, Japan, February 20-26, 2011. Proceedings, Part II, páginas 301–312.
La source de données: Dialnet