Elamur Research⌕
☰
home / papers / 2018 / Stance Detection with Hierarchical Attention Network

Stance Detection with Hierarchical Attention Network

Summary

The paper addresses stance detection, which means deciding whether a post supports or opposes a given target. Earlier neural approaches mostly built a document vector from the word sequence alone. They ignored cues that correlate with stance, such as sentiment, syntactic relations and argumentative sentences. The authors propose a Hierarchical Attention Network (HAN). It encodes the document and three linguistic views, each with its own LSTM:

  • sentiment words drawn from a lexicon,
  • word pairs connected by dependency relations,
  • argument sentences.

A linguistic attention layer links the document representation with each view. A hyper attention layer then decides how much each view contributes to the final representation. The model was tested on two benchmarks, H&N14 (debate posts) and SemEval16. HAN reaches the best averaged F1 scores against SVM, a plain LSTM and the target-aware TAN model. The study shows that explicit linguistic knowledge still benefits neural stance classifiers, especially when its weight is adjusted per document.

Key points

  • Datasets and targets:
    • H&N14 covers Abortion, GayRights, Obama and Marijuana.
    • SemEval16 covers Atheism, Climate, Feminism, Hillary and Abortion.
  • On H&N14, HAN scores 62.18 macro-F and 63.25 micro-F. TAN scores 60.49 and 62.35, and the plain LSTM scores 58.13 and 59.45.
  • On SemEval16, HAN scores 61.00 macro-F and 69.79 micro-F, compared with 59.56 and 68.79 for TAN. HAN does not win on every target: TAN is better on Climate and Hillary.
  • Each linguistic view helps on its own when added through attention. Argument information is the strongest single view, reaching 61.89 macro-F on H&N14.
  • Ablation on H&N14 lowers macro-F step by step:
    • removing hyper attention drops it to 60.95,
    • also removing linguistic attention drops it to 60.04,
    • this is still above the plain LSTM at 58.13.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Cited by

45 papers of our library cite it (of the 140673 whose bibliographies we read); 16 of them have a page here, 29 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{sun2018stance,
  title = {{Stance Detection with Hierarchical Attention Network}},
  author = {Sun, Qingying and Wang, Zhongqing and Zhu, Qiaoming and Zhou, Guodong},
  booktitle = {Proceedings of the 27th International Conference on Computational Linguistics},
  year = {2018},
  url = {https://aclanthology.org/C18-1203/}
}

References

25 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Rakesh Agrawal, Sridhar Rajagopalan, Ramakrishnan Srikant, and Yirong Xu. 2003. Mining newsgroups using networks arising from social behavior. In Proceedings of the 12th International World Wide Web Conference, pages 529–535. find
  2. Pranav Anand, Marilyn A. Walker, Rob Abbott, Jean E. Fox Tree, Robeson Bowmani, and Michael Minor. 2011. Cats rule and dogs drool!: Classifying stance in online debate. In Proceedings of the 2nd Workshop on Computational Approaches to Subjectivity and Sentiment Analysis, pages 1–9. find
  3. Isabelle Augenstein, Tim Rocktäschel, Andreas Vlachos, and Kalina Bontcheva. 2016. Stance detection with bidirectional conditional encoding. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 876–885. find our card
  4. Clinton Burfoot, Steven Bird, and Timothy Baldwin. 2011. Collective classification of congressional floor-debate transcripts. In The 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference, pages 1506–1515. find
  5. Wei-Fan Chen and Lun-Wei Ku. 2016. UTCNN: a deep learning model of stance classification on social media text. In 26th International Conference on Computational Linguistics, Proceedings of the Conference: Technical Papers, pages 1635–1645. find
  6. Jiachen Du, Ruifeng Xu, Yulan He, and Lin Gui. 2017. Stance classification with target-specific neural attention. In Proceedings of the 26th International Joint Conference on Artificial Intelligence, pages 3988–3994. find
  7. Chris Dyer, Miguel Ballesteros, Wang Ling, Austin Matthews, and Noah A. Smith. 2015. Transition-based dependency parsing with stack long short-term memory. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural… find our card
  8. Kazi Saidul Hasan and Vincent Ng. 2014. 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, pages 751–762. find
  9. Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long short-term memory. Neural Computation, 9(8):1735– 1780. find
  10. Diederik P. Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. CoRR, abs/1412.6980. find
  11. Bing Liu. 2012. Sentiment analysis and opinion mining. Synthesis lectures on human language technologies, 5(1):1–167. find
  12. Tomas Mikolov, Ilya Sutskever, Kai Chen, Gregory S. Corrado, and Jeffrey Dean. 2013. Distributed representations of words and phrases and their compositionality. In 27th Annual Conference on Neural Information Processing Systems, pages 3111–3119. find
  13. Saif Mohammad, Svetlana Kiritchenko, Parinaz Sobhani, Xiao-Dan Zhu, and Colin Cherry. 2016. Semeval-2016 task 6: Detecting stance in tweets. In Proceedings of the 10th International Workshop on Semantic Evaluation, pages 31–41. find
  14. Akiko Murakami and Rudy Raymond. 2010. Support or oppose? classifying positions in online debates from reply activities and opinion expressions. In 23rd International Conference on Computational Linguistics, pages 869–875. find
  15. Bo Pang, Lillian Lee, et al. 2008. Opinion mining and sentiment analysis. Foundations and Trends R in Information Retrieval, 2(1–2):1–135. find
  16. Isaac Persing and Vincent Ng. 2016. Modeling stance in student essays. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, pages 2174–2184. find
  17. Ashwin Rajadesingan and Huan Liu. 2014. Identifying users with opposing opinions in twitter debates. In Social Computing, Behavioral-Cultural Modeling and Prediction - 7th International Conference, pages 153–160. find
  18. Swapna Somasundaran and Janyce Wiebe. 2010. 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. find
  19. Dhanya Sridhar, James R. Foulds, Bert Huang, Lise Getoor, and Marilyn A. Walker. 2015. Joint models of disagreement and stance in online debate. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language P… find
  20. Ilya Sutskever, Oriol Vinyals, and Quoc V. Le. 2014. Sequence to sequence learning with neural networks. In Advances in Neural Information Processing Systems, pages 3104–3112. find
  21. Matt Thomas, Bo Pang, and Lillian Lee. 2006. Get out the vote: Determining support or opposition from congressional floor-debate transcripts. In Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing, pages 327–335. find
  22. Prashanth Vijayaraghavan, Soroush Vosoughi, and Deb Roy. 2016. Automatic detection and categorization of election-related tweets. In Proceedings of the 10th International Conference on Web and Social Media, pages 703–706. find
  23. Marilyn A. Walker, Pranav Anand, Rob Abbott, and Ricky Grant. 2012. Stance classification using dialogic properties of persuasion. In Human Language Technologies: Conference of the North American Chapter of the Association of Computational Linguistics, pages 592–596. find
  24. Theresa Wilson and Janyce Wiebe. 2005. Annotating attributions and private states. In Proceedings of the Workshop on Frontiers in Corpus Annotations II: Pie in the Sky, pages 53–60. find
  25. Guido Zarrella and Amy Marsh. 2016. MITRE at semeval-2016 task 6: Transfer learning for stance detection. In Proceedings of the 10th International Workshop on Semantic Evaluation, SemEval@NAACL-HLT 2016, San Diego, CA, USA, June 16-17, 2016, pages 458–463. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sun-2018-stance-detection-hierarchical-attention-network/