Aplicando Redes Neuronales Siamesas con Atención Jerárquica para resúmenes multi-documento

• Autores: Fernando García Granada , Encarna Segarra Soriano , José Ángel González, Julien Delonca, Emilio Sanchís Arnal
• Localización: Procesamiento del lenguaje natural, ISSN 1135-5948, Nº. 63, 2019, págs. 111-118
• Idioma: español
• Títulos paralelos:
  • Applying Siamese Hierarchical Attention Neural Networks for multi-document summarization
• Enlaces
  • Texto completo
• Resumen
  • español

    En este artículo presentamos una aproximación al problema de resumen automático multi-documento, basada en Redes Siamesas Jerárquico-Atencionales. El mecanismo de atención de las redes Jerárquico-Atencionales permite asignar un peso a cada frase en función de su relevancia en el proceso de clasificación. Durante la generación del resumen sólo se tienen en cuenta los pesos asociados a las frases para seleccionar aquellas más relevantes. En este trabajo exploramos la posibilidad de adaptar estos modelos al problema de resumen multi-documento (típicamente documentos muy largos donde la aplicación directa de redes neuronales no se comporta correctamente). Se ha experimentado utlizando el corpus CNN/DailyMail para entrenamiento, y el corpus DUC-2007 para evaluación. A pesar de la heterogeneidad de las características entre el corpus de entrenamiento (CNN/DailyMail) y el corpus de test (DUC-2007), los resultados muestran la adecuación de esta propuesta al resumen multi-documento.

  • English

    In this paper, we present an approach to multi-document summarization based on Siamese Hierarchical Attention Neural Networks. The attention mechanism of Hierarchical Attention Networks, provides a score to each sentence in function of its relevance in the classification process. For the summarization process, only the scores of sentences are used to rank them and select the most salient sentences. In this work we explore the adaptability of this model to the problem of multi-document summarization (typically very long documents where the straightforward application of neural networks tends to fail). The experiments were carried out using the CNN/DailyMail as training corpus, and the DUC-2007 as test corpus. Despite the difference between training set (CNN/DailyMail) and test set (DUC-2007) characteristics, the results show the adequacy of this approach to multi-document summarization. |

• Referencias bibliográficas
  • Al-Sabahi, K., Z. Zuping, and M. Nadher. 2018. A hierarchical structured self-attentive model for extractive document summarization (hssas)....
  • Begum, N., M. Fattah, and F. Ren. 2009. Automatic text summarization using support vector machine. 5:1987–1996, 07.
  • Carbonell, J. and J. Goldstein. 1998. The use of mmr, diversity-based reranking for reordering documents and producing summaries. In Proceedings...
  • Cheng, J. and M. Lapata. 2016. Neural summarization by extracting sentences and words. In Proceedings of the 54th Annual Meeting of the Association...
  • Deerwester, S., S. T. Dumais, G. W. Furnas, T. K. Landauer, and R. Harshman. 1990. Indexing by latent semantic analysis. Journal of the American...
  • Erkan, G. and D. R. Radev. 2004. Lexrank: Graph-based lexical centrality as salience in text summarization. J. Artif. Int. Res., 22(1):457–479,...
  • González, J.-Á., E. Segarra, F. GarćıaGranada, E. Sanchis, and L.-F. Hurtado. 2019. Siamese hierarchical attention networks for extractive...
  • Grusky, M., M. Naaman, and Y. Artzi. 2018. Newsroom: A dataset of 1.3 million summaries with diverse extractive strategies. In Proceedings...
  • Hochreiter, S. and J. Schmidhuber. 1997. Long short-term memory. Neural Computation, 9(8):1735–1780.
  • Lin, C.-Y. 2004. Rouge: A package for automatic evaluation of summaries. In S. S. Marie-Francine Moens, editor, Text Summarization Branches...
  • Lloret, E. and M. Palomar. 2012. Text summarisation in progress: a literature review. Artificial Intelligence Review, 37(1):1–41.
  • Nallapati, R., F. Zhai, and B. Zhou. 2017. Summarunner: A recurrent neural network based sequence model for extractive summarization of documents....
  • Nallapati, R., B. Zhou, C. N. dos Santos, Ç. Gülçehre, and B. Xiang. 2016. Abstractive text summarization using sequence-to-sequence...
  • Narayan, S., S. B. Cohen, and M. Lapata. 2018. Ranking sentences for extractive summarization with reinforcement learning. In Proceedings...
  • Paulus, R., C. Xiong, and R. Socher. 2017. A deep reinforced model for abstractive summarization. CoRR, abs/1705.04304.
  • Saggion, H. and T. Poibeau, 2013. Automatic Text Summarization: Past, Present and Future, pages 3–21. Springer Berlin Heidelberg, Berlin,...
  • See, A., P. J. Liu, and C. D. Manning. 2017. Get to the point: Summarization with pointer-generator networks. In Proceedings of the 55th Annual...
  • Shen, D., J.-T. Sun, H. Li, Q. Yang, and Z. Chen. 2007. Document summarization using conditional random fields. In Proceedings of the 20th...
  • Tur, G. and R. De Mori. 2011. Spoken language understanding: Systems for extracting semantic information from speech. John Wiley & Sons.
  • Wu, Y. and B. Hu. 2018. Learning to extract coherent summary via deep reinforcement learning. In Thirty-Second AAAI Conference on Artificial...
  • Yang, Z., D. Yang, C. Dyer, X. He, A. Smola, and E. Hovy. 2016. Hierarchical attention networks for document classification. In Proceedings...
  • Zhou, Q., N. Yang, F. Wei, S. Huang, M. Zhou, and T. Zhao. 2018. Neural document summarization by jointly learning to score and select sentences....