隨著科技快速地發展，人類不再只依賴報章雜誌或是廣播電視等傳統媒體，而是透過社群媒體獲取最新的資訊。與傳統媒體相比，社群媒體有著許多優勢，像是易於取得最新資訊，或是可使資訊迅速的傳播。然而這些優勢卻也拿來被有心人士利用，進而傳播假資訊，使得社群媒體成為影響政治或經濟的手段之一。社群媒體的特點為，人們可於任何貼文底下發表自己對於該內容的觀點或立場，透過分析這些對於特定內容的立場，可以用來偵測資訊的真實性，而立場偵測也成為目前被廣泛研究的任務。

在現行的方法中，許多研究會使用陳述句的語言特徵，或是利用注意力機制計算特定目標與陳述句的相關性，來預測陳述句的立場。然而，這些研究往往忽略了陳述句的立場並不只與該內容相關，更重要的是需要考慮特定目標的資訊。此外，陳述句的依存句法結構也總是對於特定目標表達強烈的立場。因此，我們提出了基於注意力機制與圖卷積網路的模型，該模型可以使在學習陳述句表示的過程中，同時納入特定目標以及依存句法的資訊。我們將特定目標與陳述句的內容構成一個圖，並透過圖卷積網路，使模型能學習含有與特定目標和依存句法相關資訊的陳述句表示，以獲得更全面的資訊。實驗結果證明，我們所提出的方法，確實能夠比只使用注意力機制的模型更加準確地預測立場。

With the rapid growth of technology, humans not only rely on traditional media such as newspapers or television but also through social media to get the newest information. Social media have more advantages than traditional media. For example, humans can quickly get the newest information or make disseminate information more instantly. However, someone may use these advantages to spread disinformation, which may impact political or economic. The characteristic of social media is that users can express their opinion or stance toward the post content. By analyzing these stances toward the given content, we can verify the information's correctness, and stance detection has become the most popular research task.

In current methods, most research uses two ways to classify the stance label: integrate the linguistic feature or apply the attention mechanism to calculate the correlation between the given target and claim. However, these works neglect that the stance of claim toward the target not only relates to its content but also the given target information. Moreover, the dependency relation of the claim sentence also expresses the stance toward the target. To solve this problem, we propose the model based on attention mechanism and Graph Convolutional Network (GCN) for the stance detection task. This model can integrate the target and dependency relation information into the claim representation. We build the text graph for the target and claim sentence, then learn the target-related claim representation by GCN to get comprehensive information. Experimental results show that the proposed model can classify the stance label more accurately and outperform the other baselines, which only use the attention mechanism.

[1] Pranav Anand, Marilyn Walker, Rob Abbott, Jean E Fox Tree, Robeson Bowmani, and Michael Minor. Cats rule and dogs drool!: Classifying stance in online debate. In Proceedings of the 2nd Workshop on Computational Approaches to Subjectivity and Sentiment Analysis (WASSA 2.011), pages 1–9, 2011.
[2] Leila Arras, Grégoire Montavon, Klaus­Robert Müller, and Wojciech Samek. Explaining recurrent neural network predictions in sentiment analysis. In Proceedings of the 8th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, pages 159–168, 2017.
[3] Isabelle Augenstein, Tim Rocktäschel, Andreas Vlachos, and Kalina Bontcheva. Stance detection with bidirectional conditional encoding. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 876–885, 2016.
[4] Dzmitry Bahdanau, Kyung Hyun Cho, and Yoshua Bengio. Neural machine translation by jointly learning to align and translate. In 3rd International Conference on Learning Representations, ICLR 2015, 2015.
[5] Jasmijn Bastings, Ivan Titov, Wilker Aziz, Diego Marcheggiani, and Khalil Sima＇an. Graph convolutional encoders for syntax­aware neural machine translation. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 1957–1967, 2017.
[6] Piotr Bojanowski, Édouard Grave, Armand Joulin, and Tomáš Mikolov. Enriching word vectors with subword information. Transactions of the Association for Computational Linguistics, 5:135–146, 2017.
[7] Clint Burfoot, Steven Bird, and Timothy Baldwin. Collective classification of congressional floor­-debate transcripts. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 1506–1515, 2011.
[8] Alberto Cetoli, Stefano Bragaglia, Andrew O＇Harney, and Marc Sloan. Graph convolutional networks for named entity recognition. In Proceedings of the 16th International Workshop on Treebanks and Linguistic Theories, pages 37–45, 2017.
[9] Wei­-Fan Chen and Lun­-Wei Ku. Utcnn: a deep learning model of stance classification on social media text. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pages 1635–1645, 2016.
[10] Jianpeng Cheng and Mirella Lapata. Neural summarization by extracting sentences and words. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 484–494, 2016.
[11] Kyunghyun Cho, Bart van Merriënboer, Dzmitry Bahdanau, and Yoshua Bengio. On the properties of neural machine translation: Encoder–decoder approaches. Syntax, Semantics and Structure in Statistical Translation, page 103, 2014.
[12] Kyunghyun Cho, Bart van Merriënboer, Dzmitry Bahdanau, and Yoshua Bengio. On the properties of neural machine translation: Encoder–decoder approaches. In Proceedings of SSST-­8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation, pages 103–111, 2014.
[13] Sumit Chopra, Michael Auli, and Alexander M Rush. Abstractive sentence summarization with attentive recurrent neural networks. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 93–98, 2016.
[14] Ronan Collobert, Jason Weston, Léon Bottou, Michael Karlen, Koray Kavukcuoglu, and Pavel Kuksa. Natural language processing (almost) from scratch. Journal of machine learning research, 12(ARTICLE):2493–2537, 2011.
[15] Marie­-Catherine de Marneffe, Timothy Dozat, Natalia Silveira, Katri Haverinen, Filip Ginter, Joakim Nivre, and Christopher D Manning. Universal stanford dependencies: A cross­linguistic typology. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14), pages 4585–4592, 2014.
[16] Marie-­Catherine de Marneffe, Bill MacCartney, and Christopher D Manning. Generating typed dependency parses from phrase structure parses. In Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC＇06), 2006.
[17] Kuntal Dey, Ritvik Shrivastava, and Saroj Kaushik. Topical stance detection for twitter: A two-­phase lstm model using attention. In European Conference on Information Retrieval, pages 529–536. Springer, 2018.
[18] Li Dong, Furu Wei, Chuanqi Tan, Duyu Tang, Ming Zhou, and Ke Xu. Adaptive recursive neural network for target-­dependent twitter sentiment classification. In Proceedings of the 52nd annual meeting of the association for computational linguistics (volume 2: Short papers), pages 49–54, 2014.
[19] Cicero Dos Santos and Maira Gatti. Deep convolutional neural networks for sentiment analysis of short texts. In Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: Technical Papers, pages 69–78, 2014.
[20] Jiachen Du, Ruifeng Xu, Yulan He, and Lin Gui. Stance classification with target-specific neural attention networks. In Proceedings of the 26th International Joint Conference on Artificial Intelligence, pages 3988–3994, 2017.
[21] Marco Gori, Gabriele Monfardini, and Franco Scarselli. A new model for learning in graph domains. In Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005., volume 2, pages 729–734. IEEE, 2005.
[22] Kazi Saidul Hasan and Vincent Ng. Why are you taking this stance? identifying and classifying reasons in ideological debates. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), pages 751–762, 2014.
[23] Sepp Hochreiter and Jürgen Schmidhuber. Long short-­term memory. Neural computation, 9(8):1735–1780, 1997.
[24] Qingbao Huang, Jielong Wei, Yi Cai, Changmeng Zheng, Junying Chen, Ho­-fung Leung, and Qing Li. Aligned dual channel graph convolutional network for visual question answering. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7166–7176, 2020.
[25] Zhiheng Huang, Wei Xu, and Kai Yu. Bidirectional lstm-­crf models for sequence tagging. arXiv preprint arXiv:1508.01991, 2015.
[26] Long Jiang, Mo Yu, Ming Zhou, Xiaohua Liu, and Tiejun Zhao. Target-­dependent twitter sentiment classification. In Proceedings of the 49th annual meeting of the association for computational linguistics: human language technologies, pages 151–160, 2011.
[27] Nal Kalchbrenner and Phil Blunsom. Recurrent continuous translation models. In Proceedings of the 2013 conference on empirical methods in natural language processing, pages 1700–1709, 2013.
[28] Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. In 3rd International Conference on Learning Representations, ICLR 2015, 2015.
[29] Thomas N Kipf and Max Welling. Semi­supervised classification with graph convolutional networks. 5th International Conference on Learning Representations, ICLR 2017, 2017.
[30] Guillaume Lample, Miguel Ballesteros, Sandeep Subramanian, Kazuya Kawakami, and Chris Dyer. Neural architectures for named entity recognition. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 260–270, 2016.
[31] Diego Marcheggiani and Ivan Titov. Encoding sentences with graph convolutional networks for semantic role labeling. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 1506–1515, 2017.
[32] Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781, 2013.
[33] Saif Mohammad, Svetlana Kiritchenko, Parinaz Sobhani, Xiaodan Zhu, and Colin Cherry. Semeval­-2016 task 6: Detecting stance in tweets. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval­-2016), pages 31–41, 2016.
[34] Akiko Murakami and Rudy Raymond. Support or oppose? classifying positions in online debates from reply activities and opinion expressions. In Coling 2010: Posters, pages 869–875, 2010.
[35] James W Pennebaker, Martha E Francis, and Roger J Booth. Linguistic inquiry and word count: Liwc 2001. Mahway: Lawrence Erlbaum Associates, 71(2001):2001, 2001.
[36] Jeffrey Pennington, Richard Socher, and Christopher D Manning. Glove: Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), pages 1532–1543, 2014.
[37] Minghui Qiu, Yanchuan Sim, Noah A Smith, and Jing Jiang. Modeling user arguments, interactions, and attributes for stance prediction in online debate forums. In Proceedings of the 2015 SIAM international conference on data mining, pages 855–863. SIAM, 2015.
[38] Sarvesh Ranade, Rajeev Sangal, and Radhika Mamidi. Stance classification in online debates by recognizing users＇intentions. In Proceedings of the SIGDIAL 2013 Conference, pages 61–69, 2013.
[39] Alexander M Rush, Sumit Chopra, and Jason Weston. A neural attention model for abstractive sentence summarization. In EMNLP, 2015.
[40] Franco Scarselli, Marco Gori, Ah Chung Tsoi, Markus Hagenbuchner, and Gabriele Monfardini. The graph neural network model. IEEE transactions on neural networks, 20(1):61–80, 2008.
[41] Aliaksei Severyn and Alessandro Moschitti. Twitter sentiment analysis with deep convolutional neural networks. In Proceedings of the 38th international ACM SIGIR conference on research and development in information retrieval, pages 959–962, 2015.
[42] Parinaz Sobhani, Diana Inkpen, and Stan Matwin. From argumentation mining to stance classification. In Proceedings of the 2nd Workshop on Argumentation Mining, pages 67–77, 2015.
[43] Swapna Somasundaran and Janyce Wiebe. Recognizing stances in online debates. In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP, pages 226–234, 2009.
[44] Swapna Somasundaran and Janyce Wiebe. Recognizing stances in ideological on-­line debates. In Proceedings of the NAACL HLT 2010 workshop on computational approaches to analysis and generation of emotion in text, pages 116–124, 2010.
[45] Qingying Sun, Zhongqing Wang, Qiaoming Zhu, and Guodong Zhou. Stance detection with hierarchical attention network. In Proceedings of the 27th international conference on computational linguistics, pages 2399–2409, 2018.
[46] Ilya Sutskever, Oriol Vinyals, and Quoc V Le. Sequence to sequence learning with neural networks. arXiv preprint arXiv:1409.3215, 2014.
[47] Duyu Tang, Bing Qin, Xiaocheng Feng, and Ting Liu. Effective lstms for target-dependent sentiment classification. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pages 3298–3307, 2016.
[48] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Advances in neural information processing systems, pages 5998–6008, 2017.
[49] Prashanth Vijayaraghavan, Ivan Sysoev, Soroush Vosoughi, and Deb Roy. Deepstance at semeva-l­2016 task 6: Detecting stance in tweets using character and word-­level cnns. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016), pages 413–419, 2016.
[50] Prashanth Vijayaraghavan, Soroush Vosoughi, and Deb Roy. Automatic detection and categorization of election-­related tweets. In Proceedings of the International AAAI Conference on Web and Social Media, volume 10, 2016.
[51] Duy-­Tin Vo and Yue Zhang. Target-­dependent twitter sentiment classification with rich automatic features. In Twenty­-fourth international joint conference on artificial intelligence, 2015.
[52] Marilyn Walker, Pranav Anand, Rob Abbott, and Ricky Grant. Stance classification using dialogic properties of persuasion. In Proceedings of the 2012 conference of the North American chapter of the association for computational linguistics: Human language technologies, pages 592–596, 2012.
[53] Yequan Wang, Minlie Huang, Xiaoyan Zhu, and Li Zhao. Attention-­based lstm for aspect-­level sentiment classification. In Proceedings of the 2016 conference on empirical methods in natural language processing, pages 606–615, 2016.
[54] Janyce Wiebe, Theresa Wilson, and Claire Cardie. Annotating expressions of opinions and emotions in language. Language resources and evaluation, 39(2):165–210, 2005.
[55] Theresa Ann Wilson. Fine-­grained subjectivity and sentiment analysis: recognizing the intensity, polarity, and attitudes of private states. PhD thesis, University of Pittsburgh, 2008.
[56] Chang Xu, Cecile Paris, Surya Nepal, and Ross Sparks. Cross­target stance classification with self­-attention networks. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 778–783, 2018.
[57] Jianwei Yang, Jiasen Lu, Stefan Lee, Dhruv Batra, and Devi Parikh. Graph r-­cnn for scene graph generation. In Proceedings of the European conference on computer vision (ECCV), pages 670–685, 2018.
[58] Liang Yao, Chengsheng Mao, and Yuan Luo. Graph convolutional networks for text classification. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 7370–7377, 2019.
[59] Guido Zarrella and Amy Marsh. Mitre at semeval-­2016 task 6: Transfer learning for stance detection. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-­2016), pages 458–463, 2016.
[60] Yuhao Zhang, Peng Qi, and Christopher D Manning. Graph convolution over pruned dependency trees improves relation extraction. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2205–2215, 2018.
[61] Yiwei Zhou, Alexandra I Cristea, and Lei Shi. Connecting targets to tweets: Semantic attention­based model for target-­specific stance detection. In International Conference on Web Information Systems Engineering, pages 18–32. Springer, 2017.