Elamur Research⌕
☰
home / papers / 2020 / Enhancing Cross-target Stance Detection with Transferable Semantic-Emotion Knowledge

Enhancing Cross-target Stance Detection with Transferable Semantic-Emotion Knowledge

Summary

Stance detection decides whether a text supports, opposes, or is neutral toward a given target. Labeled data for each new target is expensive to collect. This paper addresses the cross-target setting, where a classifier trained on one target must handle another. The authors argue that earlier transfer methods, which rely on shared words or latent topics, struggle for two reasons: social-media posts are short and informal, and stances are often expressed implicitly. Their model, SEKT, instead uses external lexical knowledge as a target-independent link. It builds a heterogeneous graph whose nodes are SenticNet words and EmoLex emotion tags, and encodes k-hop subgraphs with a two-layer GCN. The resulting embeddings feed a BiLSTM whose cell gains an extra knowledge-aware memory unit (KAMU), which controls how much lexicon knowledge shapes each hidden state. On an expanded version of the SemEval-2016 Task 6 data, SEKT outperforms earlier cross-target systems and alternative knowledge-fusion variants in most source→destination pairs. This suggests that lexicon-derived semantic and affective knowledge can reduce annotation needs for new topics.

Key points

  • Graph construction: In the SE-graph, edges link SenticNet words to semantically related words and to EmoLex emotion tags, weighted by co-occurrence. Word–tag weights are scaled down by a constant so that emotion concepts, which have many connections, do not dominate.
  • Target conditioning: The target is encoded by its own BiLSTM. Its final state initializes the sentence BiLSTM, producing a target-aware sentence representation.
  • Main gains: In F1avg, SEKT improves over the strongest baseline by 5.7% on FM→LA, 3.5% on LA→FM and 5.5% on DT→HC. Simply concatenating knowledge with the context representation (CrossNet-C) performed worst.
  • Ablation and hops: Removing either the SE-graph or KAMU lowers scores. GCN depth works best at 2–3 hops; 4 hops adds noise (e.g., LA→FM falls from 0.513 to 0.369).
  • Error analysis: Among 100 misclassified tweets, errors come from opinions that require world knowledge absent from the lexicons (e.g., NBC vs. Donald Trump) and from hashtags with hidden meanings.

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

42 papers of our library cite it (of the 140673 whose bibliographies we read); 13 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. DOI 10.18653/v1/2020.acl-main.291

@inproceedings{zhang2020enhancing,
  title = {{Enhancing Cross-target Stance Detection with Transferable Semantic-Emotion Knowledge}},
  author = {Zhang, Bowen and Yang, Min and Li, Xutao and Ye, Yunming and Xu, Xiaofei and Dai, Kuai},
  booktitle = {Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics},
  year = {2020},
  doi = {10.18653/v1/2020.acl-main.291},
  url = {https://doi.org/10.18653/v1/2020.acl-main.291}
}

References

23 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. I Augenstein, T Rocktaeschel, A Vlachos, and K Bontcheva. 2016. Stance detection with bidirectional conditional encoding. In Proceedings of the Conference on Empirical Methods in Natural Language Processing. Sheffield. find our card
  2. Erik Cambria, Soujanya Poria, Devamanyu Hazarika, and Kenneth Kwok. 2018. Senticnet 5: Discovering conceptual primitives for sentiment analysis by means of context embeddings. In Thirty-Second AAAI Conference on Artificial Intelligence. find
  3. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,… find
  4. Kuntal Dey, Ritvik Shrivastava, and Saroj Kaushik. 2018. Topical stance detection for twitter: A two-phase lstm model using attention. In European Conference on Information Retrieval, pages 529–536. Springer. find
  5. Mauro Dragoni and Giulio Petrucci. 2018. A fuzzy-based strategy for multi-domain sentiment analysis. International Journal of Approximate Reasoning, 93:59–73. find
  6. Jiachen Du, Ruifeng Xu, Yulan He, and Lin Gui. 2017. Stance classification with target-specific neural attention networks. International Joint Conferences on Artificial Intelligence. find
  7. Zhiting Hu, Xuezhe Ma, Zhengzhong Liu, Eduard Hovy, and Eric Xing. 2016. Harnessing deep neural networks with logic rules. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2410–2420. find
  8. Yoon Kim. 2014. Convolutional neural networks for sentence classification. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1746–1751. find our card
  9. Thomas N Kipf and Max Welling. 2016. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907. find our card
  10. Zeyang Lei, Yujiu Yang, and Min Yang. 2018. SAAN: A sentiment-aware attention network for sentiment analysis. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, SIGIR 2018, Ann Arbor, MI, USA, July 08-12, 2018, pages 1197–1200. ACM. find
  11. Yukun Ma, Haiyun Peng, and Erik Cambria. 2018. Targeted aspect-based sentiment analysis via embedding commonsense knowledge into an attentive lstm. In Thirty-Second AAAI Conference on Artificial Intelligence. find
  12. Katerina Margatina, Christos Baziotis, and Alexandros Potamianos. 2019. Attention-based conditioning methods for external knowledge integration. arXiv preprint arXiv:1906.03674. find
  13. Saif M Mohammad and Peter D Turney. 2013. Crowdsourcing a word–emotion association lexicon. Computational Intelligence, 29(3):436–465. find
  14. Swapna Somasundaran and Janyce Wiebe. 2009. 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: Volume 1-Volume 1, pages 226–234. Association for Co… find
  15. Qingying Sun, Zhongqing Wang, Qiaoming Zhu, and Guodong Zhou. 2018. Stance detection with hierarchical attention network. In Proceedings of the 27th International Conference on Computational Linguistics, pages 2399–2409. find our card
  16. 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. Association for Computational Linguistics. find
  17. Marilyn A Walker, Pranav Anand, Robert Abbott, and Ricky Grant. 2012. 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. Assoc… find
  18. Penghui Wei, Junjie Lin, and Wenji Mao. 2018. Multi-target stance detection via a dynamic memory-augmented network. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, pages 1229–1232. ACM. find
  19. Penghui Wei and Wenji Mao. 2019. Modeling transferable topics for cross-target stance detection. In Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 1173–1176. ACM. find
  20. Chang Xu, Cecile Paris, Surya Nepal, and Ross Sparks. 2018. Cross-target 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. find
  21. Bowen Zhang, Xiaofei Xu, Min Yang, Xiaojun Chen, and Yunming Ye. 2018. Cross-domain sentiment classification by capsule network with semantic rules. IEEE Access, 6:58284–58294. find
  22. Chen Zhang, Qiuchi Li, and Dawei Song. 2019a. Aspect-based sentiment classification with aspect-specific graph convolutional networks. arXiv preprint arXiv:1909.03477. find
  23. Jingqing Zhang, Piyawat Lertvittayakumjorn, and Yike Guo. 2019b. Integrating semantic knowledge to tackle zero-shot text classification. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long a… find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/zhang-2020-enhancing-cross-target-stance-detection/