基于篇章结构图网络的话题分割

Abstract
Figure/Table
References (0)
Related Citation (4)

Download: PDF (5175 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract Topic segmentation, as one of the classic tasks in the field of natural language processing, is to segment the input discourse into paragraphs with continuous semantics. Previous works used word frequencybased, latentbased, sequentialbased, and Transformerbased methods to encode sentences, ignoring modeling global semantic information of the discourse. This paper proposes to use Discourse Structure Graph Network to encode sentences for a sentence representation with the global information of the discourse. In detail, the model firstly constructs a single graph for each discourse, which contains all sentences and word nodes of it as well as the adjacency information between them. The model then uses Gated Graph Neural Networks to iterate the graph that gets the sentence representation with the global information of the discourse. They are finally fed to the Bi-LSTM layer to predict the segmentation points. The experimental results demonstrate that the model gets a more suitable sentence representation than other baselines for topic segmentation and achieves the best performance on various popular datasets.

Key words： topic segmentation sentence encoding graph neural networks

Received: 17 March 2021

ZTFLH:

TP391

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	XU Shaoyang
	JIANG Feng
	LI Peifeng

Cite this article:

XU Shaoyang,JIANG Feng,LI Peifeng. Topic Segmentation via Discourse Structure Graph Network[J]. , 2021, 35(12): 17-27.

URL:

http://jcip.cipsc.org.cn/EN/ OR http://jcip.cipsc.org.cn/EN/Y2021/V35/I12/17

［1］Tur G, De Mori R. Spoken language understanding: Systems for extracting semantic information from speech［M］. New Jersey: John Wiley and Sons, 2011.
［2］Koshorek O, Cohen A, Mor N, et al. Text segmentation as a supervised learning task［C］//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2018: 469-473.
［3］Hearst M A. TextTiling: segmenting text into multi-paragraph subtopic passages［J］. Computational Linguistics, 1997, 23(1): 33-64.
［4］Dennis S, Landauer T, Kintsch W, et al. Introduction to latent semantic analysis［C］//Proceedings of the 25th Annual Meeting of the Cognitive Science Society, 2003: 25.
［5］Choi F Y Y. Advances in domain independent linear text segmentation［C］//Proceedings of the 6th Applied Natural Language Processing Conference, 2000: 26-33.
［6］Glava G, Nanni F, Ponzetto S P. Unsupervised text segmentation using semantic relatedness graphs［C］//Proceedings of the 5th Joint Conference on Lexical and Computational Semantics, 2016.
［7］Blei D M, Ng A Y, Jordan M I. Latent dirichlet allocation［J］. The Journal of Machine Learning Research, 2003, 3: 993-1022.
［8］Barrow J, Jain R, Morariu V, et al. A joint model for document segmentation and segment labeling［C］//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 313-322.
［9］Glava G, Somasundaran S. Two-level transformer and auxiliary coherence modeling for improved text segmentation［C］//Proceedings of the 34th AAAI Conference on Artificial Intelligence, 2020: 2306-2315.
［10］Badjatiya P, Kurisinkel L J, Gupta M, et al. Attention-based neural text segmentation［C］//Proceedings of the European Conference on Information Retrieval, 2018: 180-193.
［11］Li J, Sun A, Joty S R. SegBot: A generic neural text segmentation model with pointer network［C］//Proceedings of the International Joint Conference on Artificial Intelligence, 2018: 4166-4172.
［12］Lukasik M, Dadachev B, Simoes G, et al. Text segmentation by cross segment attention［C］//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2020: 4707-4716.
［13］Xing L, Hackinen B, Carenini G, et al. Improving context modeling in neural topic segmentation［C］//Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing, 2020: 626-636.
［14］Tai K S, Socher R, Manning C D. Improved semantic representations from tree-structured long short-term memory networks［C］//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguis, 2015: 1556-1566.
［15］Shi H, Zhou H, Chen J, et al. On tree-based neural sentence modeling［C］//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2018: 4631-4641.
［16］Vaswani A, Shazeern N, Parmar N, et al. Attention is all you need［C］//Proceedings of the 31st International Conference on Neural Information Processing Systems, 2017: 5998-6008.
［17］Yao L, Mao C, Luo Y. Graph convolutional networks for text classification［C］//Proceedings of the 33rd AAAI Conference on Artificial Intelligence, 2019: 7370-7377.
［18］Zhang Y, Yu X, Cui Z, et al. Every document owns its structure: Inductive text classification via graph neural networks［C］//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 334-339.
［19］Li Y, Tarlow D, Brockschmidt M, et al. Gated graph sequence neural networks［C］//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 273-283.
［20］Pennington J, Socher R, Manning C D. GloVe: Global vectors for word representation［C］//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2014: 1532-1543.
［21］Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks［C］//Proceedings of the 5th International Conference on Learning Representations, 2017.
［22］Huang L, Ma D, Li S, et al. Text level graph neural network for text classification［C］//Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, 2019: 3442-3448.
［23］Sun Z, Jiang F, Li P, et al. Macro discourse relation recognition via discourse argument pair graph［C］//Proceedings of the 2020 CCF International Conference on Natural Language Processing and Chinese Computing, 2020: 108-119.
［24］Arnold S, Schneider R, Cudré Mauroux P, et al. Sector: A neural model for coherent topic segmentation and classification［J］. Transactions of the Association for Computational Linguistics, 2019, 7: 169-184.
［25］Malioutov I, Barzilay R. Minimum cut model for spoken lecture segmentation［C］//Proceedings of the 44th Annual Meeting of the Association for Computational Linguistics, 2006: 25-32.
［26］Chen H, Branavan S R K, Barzilay R, et al. Global models of document structure using latent permutations［C］//Proceedings of the North American Chapter of the Association of Computational Linguistics, 2009: 371-379.
［27］Beeferman D, Berger A, Lafferty J. Statistical models for text segmentation［J］. Machine Learning, 1999, 34(1): 177-210.
［28］Kingma D P, Ba J. Adam: A method for stochastic optimization［C］//Proceedings of the 3rd International Conference on Learning Representation, 2014.