Similar articles

Abstract: A lack of labeled corpora obstructs the research progress on implicit discourse relation recognition (DRR) for Chinese, while there are some available discourse corpora in other languages, such as English. In this paper, we propose a cross-lingual implicit DRR framework that exploits an available English corpus for the Chinese DRR task. We use machine translation to generate Chinese instances from a labeled English discourse corpus. In this way, each instance has two independent views: Chinese and English views. Then we train two classifiers in Chinese and English in a co-training way, which exploits unlabeled Chinese data to implement better implicit DRR for Chinese. Experimental results demonstrate the effectiveness of our method.

基于协同学习的跨语言隐式篇章关系识别

[1]Biran O, McKeown K, 2013. Aggregated word pair features for implicit discourse relation disambiguation. Proc 51^st Annual Meeting of the Association for Computational Linguistics, p.69-73.

[2]Blum A, Mitchell T, 1998. Combining labeled and unlabeled data with cotraining. Proc 11^th Annual Conf on Computational Learning Theory, p.92-100.

[3]Braud C, Denis P, 2015. Comparing word representations for implicit discourse relation classification. Proc Conf on Empirical Methods in Natural Language Processing, p.2201-2211.

[4]Carlson L, Marcu D, Okurowski M, 2001. Building a discourse-tagged corpus in the framework of rhetorical structure theory. Proc 2^nd SIGDIAL Workshop on Discourse and Dialogue, p.1-10.

[5]Chen J, Zhang Q, Liu P, et al., 2016. Implicit discourse relation detection via a deep architecture with gated relevance network. Proc 54^th Annual Meeting of the Association for Computational Linguistics, p.1726-1735.

[6]Chen L, 2006. English and Chinese Discourse Structure Dimension Theory and Practice. PhD Thesis, Shanghai International Studies University, China.

[7]Chiarcos C, 2012. Towards the unsupervised acquisition of discourse relations. Proc 50^th Annual Meeting of the Association for Computational Linguistics, p.213-217.

[8]Cimiano P, Reyle U, Šarić J, 2005. Ontology-driven discourse analysis for information extraction. Data Knowl Eng, 55(1):59-83.

[8]Clark S, Curran J, Osborne M, 2003. Bootstrapping POS-taggers using unlabelled data. Proc 7^th Conf on Natural Language Learning, p.49-55.

[10]Guzmán F, Joty S, Màrquez L, et al., 2014. Using discourse structure improves machine translation evaluation. Proc 52^nd Annual Meeting of the Association for Computational Linguistics, p.687-698.

[11]Hernault H, Bollegala D, Ishizuka M, 2010. Towards semi-supervised classification of discourse relations using feature correlations. Proc SIGDIAL Conf and the 11^th Annual Meeting of the Special Interest Group on Discourse and Dialogue, p.55-58.

[12]Huang H, Chen H, 2011. Chinese discourse relation recognition. Proc 5^th Int Joint Conf on Natural Language Processing, p.1442-1446.

[13]Jain S, Batra S, 2015. Cross lingual sentiment analysis using modified BRAE. Proc Conf on Empirical Methods in Natural Language Processing, p.159-168.

[14]Ji Y, Eisenstein J, 2015. One vector is not enough: entity-augmented distributed semantics for discourse relations. Trans Assoc Comput Ling, 3:329-344.

[15]Ji Y, Haffari G, Eisenstein J, 2016. A latent variable recurrent neural network for discourse relation language models. Proc Conf North American Chapter of the Association for Computational Linguistics on Human Language Technologies, p.332-342.

[16]Laali M, Kosseim L, 2014. Inducing discourse connectives from parallel texts. Proc 25^th Int Conf on Computational Linguistics, p.610-619.

[17]Lan M, Xu Y, Niu Z, 2013. Leveraging synthetic discourse data via multi-task learning for implicit discourse relation recognition. Proc 51^st Annual Meeting of the Association for Computational Linguistics, p.476-485.

[18]Li J, Carpuat M, Nenkova A, 2014. Cross-lingual discourse relation analysis: a corpus study and a semi-supervised classification system. Proc 25^th Int Conf on Computational Linguistics, p.577-587.

[19]Li Y, Feng W, Sun J, et al., 2014. Building Chinese discourse corpus with connective-driven dependency tree structure. Proc Conf on Empirical Methods in Natural Language Processing, p.2105-2114.

[20]Lin Z, Kan M, Ng H, 2009. Recognizing implicit discourse relations in the Penn discourse treebank. Proc Conf on Empirical Methods in Natural Language Processing, p.343-351.

[21]Liu Y, Li S, Zhang X, et al., 2016. Implicit discourse relation classification via multi-task neural networks. Proc 30^th Conf on Artificial Intelligence, p.2750-2756.

[22]Louis A, Joshi A, Prasad R, et al., 2010. Using entity features to classify implicit discourse relations. Proc 11^th Annual Meeting of the Special Interest Group on Discourse and Dialogue, p.59-62.

[23]Marcu D, Echihabi A, 2002. An unsupervised approach to recognizing discourse relations. Proc 40^th Annual Meeting of the Association for Computational Linguistics, p.368-375.

[24]Miltsakaki E, Dinesh N, Prasad R, et al., 2005. Experiments on sense annotations and sense disambiguation of discourse connectives. Proc 4^th Workshop on Treebanks and Linguistic Theories, p.1-13.

[25]Ming Y, 2008. Rhetorical structure annotation of Chinese news commentaries. J Chin Inform Proc, 22(4):19-23.

[26]Ng V, Cardie C, 2003. Weakly supervised natural language learning without redundant views. Proc Conf North American Chapter of the Association for Computational Linguistics on Human Language Technology, p.94-101.

[27]Park J, Cardie C, 2012. Improving implicit discourse relation recognition through feature set optimization. Proc 13^th Annual Meeting of the Special Interest Group on Discourse and Dialogue, p.108-112.

[28]Pitler E, Nenkova A, 2009. Using syntax to disambiguate explicit discourse connectives in text. Proc ACL-IJCNLP Conf, p.13-16.

[29]Pitler E, Louis A, Nenkova A, 2009. Automatic sense prediction for implicit discourse relations in text. Proc of the Joint Conf 47^th Annual Meeting of the ACL and the 4^th Int Joint Conf on Natural Language Processing of the AFNLP, p.683-691.

[30]Prasad R, Dinesh N, Lee A, et al., 2008. The Penn discourse treebank 2.0. Proc Int Conf on Language Resources and Evaluation, p.2961-2968.

[31]Qian L, Hui H, Hu Y, et al., 2014. Bilingual active learning for relation classification via pseudo parallel corpora. Proc 52^nd Annual Meeting of the Association for Computational Linguistics, p.582-592.

[32]Qin L, Zhang Z, Zhao H, 2016. A stacking gated neural architecture for implicit discourse relation classification. Proc Conf on Empirical Methods in Natural Language Processing, p.2263-2270.

[33]Rutherford A, Xue N, 2014. Discovering implicit discourse relations through brown cluster pair representation and coreference patterns. Proc 14^th Conf European Chapter of the Association for Computational Linguistics, p.645-654.

[34]Rutherford A, Xue N, 2015. Improving the inference of implicit discourse relations via classifying explicit discourse connectives. Proc Conf of the North American Chapter of the Association for Computational Linguistics on Human Language Technologies, p.799-808.

[35]Rutherford A, Demberg V, Xue N, 2016. Neural network models for implicit discourse relation classification in English and Chinese without surface features. http://arxiv.org/abs/1606.01990

[36]Sarkar A, 2001. Applying co-training methods to statistical parsing. Proc 2^nd Meeting of the North American Chapter of the Association for Computational Linguistics on Language Technologies, p.1-8.

[37]Sporleder C, Lascarides A, 2008. Using automatically labelled examples to classify rhetorical relations: an assessment. Nat Lang Eng, 14:369-416.

[38]Verberne S, Boves L, Oostdijk N, et al., 2007. Evaluating discourse-based answer extraction for why-question answering. Proc 30^th Annual Int ACM SIGIR Conf on Research and Development in Information Retrieval, p.735-736.

[39]Wan X, 2009. Co-training for cross-lingual sentiment classification. Proc of the Joint Conf 47^th Annual Meeting of the ACL and the 4^th Int Joint Conf on Natural Language Processing of the AFNLP, p.235-243.

[40]Wang W, Su J, Tan C, 2010. Kernel based discourse relation recognition with temporal ordering information. Proc 48^th Annual Meeting of the Association for Computational Linguistics, p.710-719.

[41]Wang X, Li S, Li J, et al., 2012. Implicit discourse relation recognition by selecting typical training examples. Proc of COLING, p.2757-2772.

[42]Xue N, 2005. Annotating discourse connectives in the Chinese treebank. Proc Workshop on Frontiers in Corpus Annotations II: Pie in the Sky, p.84-91.

[43]Zhang B, Su J, Xiong D, et al., 2015. Shallow convolutional neural network for implicit discourse relation recognition. Proc Conf on Empirical Methods in Natural Language Processing, p.2230-2235.

[44]Zhang B, Xiong D, Su J, et al., 2016. Variational neural discourse relation recognizer. Proc Conf on Empirical Methods in Natural Language Processing, p.382-391.

[45]Zhang M, Song Y, Qin B, et al., 2013. Chinese discourse relation recognition. J Chin Inform Proc, 27(6):51-57.

[46]Zhang M, Qin B, Liu T, 2014. Chinese discourse relation hierarchy and annotation. J Chin Inform Proc, 28(2):28-36.

[47]Zhou L, Gao W, Li B, et al., 2012. Cross-lingual identification of ambiguous discourse connectives for resource poor language. Proc COLING, p.1409-1418.

[48]Zhou Y, Xue N, 2015. The Chinese discourse treebank: a Chinese corpus annotated with discourse relations. Lang Res Eval, 49(2):397-431.

[49]Zhou Z, Xu Y, Niu Z, et al., 2010. Predicting discourse connectives for implicit discourse relation recognition. 23^rd Int Conf on Computational Linguistics, p.1507-1514.