Elamur Research⌕
☰
home / papers / 2020 / Token Sequence Labeling vs. Clause Classification for English Emotion Stimulus Detection

Token Sequence Labeling vs. Clause Classification for English Emotion Stimulus Detection

Paper · 2020 · Laura Ana Maria Oberländer and Roman Klinger

Summary

The paper asks which unit to predict when locating the cause (stimulus) of an emotion in English text. Mandarin research mostly classifies whole clauses as stimulus or not. English work has mostly tagged individual tokens. The two styles had never been compared directly. The authors build one framework that runs both formulations side by side. It includes a clause segmenter for English, based on constituency parses from the Berkeley Neural Parser, which supplies clauses when none are annotated. Both formulations use nearly identical neural models (embeddings, BiLSTM, attention), and only the output layer differs. Mapping rules convert each kind of output into the other, so both can be evaluated in terms of whole clauses as well as individual tokens. On four English datasets from different domains, token labeling wins on three, under both evaluation views. Clause classification wins only on the dataset with a high density of clause annotations. An error analysis shows that English stimuli often do not line up with clause boundaries. The code and the clause and token annotations are released.

Key points

  • Three models are compared:
    • SL: token-level IOB tagging with GloVe embeddings, a BiLSTM, attention and a CRF output layer.
    • ICC: each clause is classified independently.
    • JCC: clauses are classified jointly.
  • The clause extractor works as follows:
    • It cuts the constituency tree at the borders of S, SBAR, SBARQ, INV and SQ nodes.
    • It then repeatedly merges segments made only of punctuation into the previous segment.
    • It merges segments of 3 or fewer tokens into the next segment.
  • Outputs are mapped in both directions:
    • Tokens to clauses: a clause counts as a stimulus if any of its tokens is labeled B or I.
    • Clauses to tokens: in a stimulus clause, the first token becomes B and the rest become I; all other tokens get O.
  • The datasets include EmotionStimulus, ElectoralTweets, EmotionCauseAnalysis and a headlines corpus.
    • ElectoralTweets, which was crowdsourced, gave the weakest SL results.
    • The well-formed sentences of EmotionStimulus were the easiest.
  • Typical errors:
    • Predicted spans stop at coordinating conjunctions and miss the second conjunct.
    • Spans sometimes swallow the emotion cue word itself.

Interesting ideas

  • I1. When comparing two task formulations, keep the encoder identical and change only the output layer, so that any performance gap reflects the formulation rather than the architecture.
  • I2. Define a deterministic mapping in both directions between coarse units and token labels, so that systems of different granularity can be scored under both views.
  • I3. Build a cheap clause segmenter from parse-tree node labels, then repair it with simple merge rules for punctuation-only and very short fragments.
  • I4. The finding that English emotion causes cross clause boundaries suggests treating the stimulus as a flexible argument span, much like a semantic role, rather than a fixed syntactic unit.
  • I5. The error pattern where spans stop at coordinating conjunctions points to an easy rule-level fix: extend a span over conjoined siblings.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

  • I1 (learn): Use this controlled design when deciding between span-level and constituent-level targets for any extractor, so that results are not confounded by model capacity.
  • I2 (learn): Score frame-argument extraction both as exact token spans and as "touches the right constituent", so the evaluation shows whether errors come from boundaries or from detection.
  • I3 (tree): Derive clause units from the UD tree (for example, from the heads of advcl, ccomp and conj relations) and add similar merge heuristics, giving an explainable segmentation layer for later stages.
  • I4 (frames): Model the stimulus as a role of emotion frames in prag (similar to a VerbNet Stimulus), with its span read off the dependency subtree instead of the clause.
  • I5 (frames): Add an explicit rule that propagates a role span across conj dependents, so that "the loss and the pain" is captured as a whole.

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

 Institutions

🎓University of StuttgartStuttgart · Germany75 works · 46,113 citations

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 9,086 worksArtificial Intelligence · 8,889 worksArts and Humanities · 9,296 worksLanguage and Linguistics · 9,206 worksPsychology · 798 worksExperimental and Cognitive Psychology · 463 works

🧭 Topics 2Research subjects and languages the work is about.English · 1,373 worksSentiment & emotion · 608 works

Cite

Cite as Laura Ana Maria Oberländer and Roman Klinger. 2020. Token Sequence Labeling vs. Clause Classification for English Emotion Stimulus Detection. Proceedings of the Ninth Joint Conference on Lexical and Computational Semantics.

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

@inproceedings{oberlander2020token,
  title = {{Token Sequence Labeling vs. Clause Classification for English Emotion Stimulus Detection}},
  author = {Laura Ana Maria Oberländer and Roman Klinger},
  booktitle = {Proceedings of the Ninth Joint Conference on Lexical and Computational Semantics},
  year = {2020},
  url = {https://aclanthology.org/2020.starsem-1.7/}
}

References

46 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Song Feng, Ritwik Banerjee, and Yejin Choi. 2012. Characterizing stylistic elements in syntactic structure. In Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 1522–1533, Jeju Island, Korea. Association for Computational Linguistics. find
  2. Ann Bies, Mark Ferguson, Karen Katz, Robert Mac-Intyre, Victoria Tredinnick, Grace Kim, Mary Ann Marcinkiewicz, and Britta Schasberger. 1995. Bracketing guidelines for Treebank II style Penn Treebank project. Online: http://languagelog.ldc.upenn.edu/myl/PennTreebank1995.pdf. ↗
  3. Gayatree Ganu, Noemie Elhadad, and Amélie Marian. 2009. Beyond the stars: Improving rating predictions using review text content. In Twelfth International Workshop on the Web and Databases (WebDB 2009). find
  4. Laura Ana Maria Bostan, Evgeny Kim, and Roman Klinger. 2020. GoodNewsEveryone: A corpus of news headlines annotated with emotions, semantic roles, and reader perception. In Proceedings of The 12th Language Resources and Evaluation Conference, pages 1554–1566, Marseille, France. European Language Resources Association. ↗
  5. Kai Gao, Hua Xu, and Jiushuo Wang. 2015. A rule-based approach to emotion cause detection for chinese micro-blogs. Expert Systems with Applications, 42(9):4517–4528. ↗
  6. Ying Chen, Wenjun Hou, Xiyao Cheng, and Shoushan Li. 2018. Joint learning for emotion classification and emotion cause detection. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 646–651, Brussels, Belgium. Association for Computational Linguistics. find
  7. Qinghong Gao, Jiannan Hu, Ruifeng Xu, Gui Lin, Yulan He, Qin Lu, and Kam-Fai Wong. 2017. Overview of NTCIR-13 ECA task. In Proceedings of the 13th NTCIR Conference on Evaluation of Information Access Technologies, pages 361–366, Tokyo, Japan. find
  8. Ying Chen, Sophia Yat Mei Lee, Shoushan Li, and Chu-Ren Huang. 2010. Emotion cause detection with linguistic constructions. In Proceedings of the 23rd International Conference on Computational Linguistics (Coling 2010), pages 179–187, Beijing, China. Coling 2010 Organizing Committee. find
  9. Matt Gardner, Joel Grus, Mark Neumann, Oyvind Tafjord, Pradeep Dasigi, Nelson F. Liu, Matthew Peters, Michael Schmitz, and Luke Zettlemoyer. 2018. AllenNLP: A deep semantic natural language processing platform. In Proceedings of Workshop for NLP Open Source Software (NLP-OSS), pages 1– 6, Melbourne, Australia. Association for Computational Linguistics.
  10. Xiyao Cheng, Ying Chen, Bixiao Cheng, Shoushan Li, and Guodong Zhou. 2017. An emotion cause corpus for chinese microblogs with multiple-user structures. ACM Transactions on Asian and Low-Resource Language Information Processing, 17(1):6:1–6:19. ↗
  11. Diman Ghazi, Diana Inkpen, and Stan Szpakowicz. 2015. Detecting emotion stimuli in emotion-bearing sentences. In International Conference on Intelligent Text Processing and Computational Linguistics, pages 152–165. Springer. find
  12. Zixiang Ding, Huihui He, Mengran Zhang, and Rui Xia. 2019. From independent prediction to reordered prediction: Integrating relative position and global label information to emotion cause identification. In Proceedings of the AAAI Conference on Artificial Intelligence, pages 6343–6350. find
  13. Lin Gui, Jiannan Hu, Yulan He, Ruifeng Xu, Qin Lu, and Jiachen Du. 2017. A question answering approach for emotion cause extraction. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 1593–1602, Copenhagen, Denmark. Association for Computational Linguistics. find
  14. Paul Ekman. 1992. An argument for basic emotions. Cognition & emotion, 6(3-4):169–200. ↗
  15. Chuang Fan, Hongyu Yan, Jiachen Du, Lin Gui, Lidong Bing, Min Yang, Ruifeng Xu, and Ruibin Mao. 2019. A knowledge regularized hierarchical approach for emotion cause analysis. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 5614–5624, Hong Kong, China. Association for Computational Linguistics. find
  16. Lin Gui, Dongyin Wu, Ruifeng Xu, Qin Lu, and Yu Zhou. 2016. Event-driven emotion cause extraction with corpus construction. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 1639–1649, Austin, Texas. Association for Computational Linguistics. ↗
  17. Niklas Jakob and Iryna Gurevych. 2010. Extracting opinion targets in a single and cross-domain setting with conditional random fields. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 1035–1045. Association for Computational Linguistics.
  18. Chuang Fan, Chaofa Yuan, Jiachen Du, Lin Gui, Min Yang, and Ruifeng Xu. 2020. Transition-based directed graph construction for emotion-cause pair extraction. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 3707–3717, Online. Association for Computational Linguistics. find
  19. Evgeny Kim and Roman Klinger. 2018. Who feels what and why? annotation of a literature corpus Japan. Asian Federation of Natural Language Processing. with semantic roles of emotions. In Proceedings of the 27th International Conference on Computational Linguistics, pages 1345–1359. Association for Computational Linguistics. find
  20. Jeffrey Pennington, Richard Socher, and Christopher Manning. 2014. Glove: Global vectors for word representation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1532–1543, Doha, Qatar. Association for Computational Linguistics. ↗
  21. Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings. ↗
  22. Robert Plutchik. 2001. The nature of emotions human emotions have deep evolutionary roots, a fact that may explain their complexity and provide tools for clinical practice. American Scientist, 89(4):344– 350. find
  23. Nikita Kitaev and Dan Klein. 2018. Constituency parsing with a self-attentive encoder. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2676–2686, Melbourne, Australia. Association for Computational Linguistics. ↗
  24. Maria Pontiki, Dimitris Galanis, Haris Papageorgiou, Ion Androutsopoulos, Suresh Manandhar, Mohammad AL-Smadi, Mahmoud Al-Ayyoub, Yanyan Zhao, Bing Qin, Orphée De Clercq, Véronique Hoste, Marianna Apidianaki, Xavier Tannier, Natalia Loukachevitch, Evgeniy Kotelnikov, Nuria Bel, Salud Marı́a Jiménez-Zafra, and Gülşen Eryiğit. 2016. SemEval-2016 task 5: Aspect based sentiment analysis. In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016), pages 19–30, San Diego, California. Association for Computational Linguistics. ↗
  25. Roman Klinger and Philipp Cimiano. 2013. Bidirectional inter-dependencies of subjective expressions and targets and their value for a joint model. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 848–854, Sofia, Bulgaria. Association for Computational Linguistics. find
  26. Paul R. Kroeger. 2005. Analyzing grammar: An introduction. Cambridge University Press. find
  27. John Lafferty, Andrew McCallum, and Fernando Pereira. 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In International Conference on Machine Learning, pages 282–289. find
  28. Maria Pontiki, Dimitris Galanis, Haris Papageorgiou, Suresh Manandhar, and Ion Androutsopoulos. 2015. SemEval-2015 task 12: Aspect based sentiment analysis. In Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015), pages 486–495, Denver, Colorado. Association for Computational Linguistics.
  29. Sophia Yat Mei Lee, Ying Cohen, Shoushan Li, and Chu-Ren Huang. 2010. Emotion cause events: Corpus construction and analysis. In Proceedings of the Seventh International Conference on Language Resources and Evaluation (LREC’10), pages 1121–1128, Valletta, Malta. European Language Resources Association (ELRA). find
  30. Vasin Punyakanok, Dan Roth, and Wen-tau Yih. 2008. The importance of syntactic parsing and inference in semantic role labeling. Computational Linguistics, 34(2):257–287. ↗
  31. Lance Ramshaw and Mitch Marcus. 1995. Text chunking using transformation-based learning. In Third Workshop on Very Large Corpora, pages 82–94.
  32. Weiyuan Li and Hua Xu. 2014. Text-based emotion classification using emotion cause extraction. Expert Systems with Applications, 41(4):1742–1749. ↗
  33. Irene Russo, Tommaso Caselli, Francesco Rubino, Ester Boldrini, and Patricio Martı́nez-Barco. 2011. EMOCause: An easy-adaptable approach to extract emotion cause contexts. In Proceedings of the 2nd Workshop on Computational Approaches to Subjectivity and Sentiment Analysis (WASSA 2.011), pages 153–160, Portland, Oregon. Association for Computational Linguistics. find
  34. Xiangju Li, Kaisong Song, Shi Feng, Daling Wang, and Yifei Zhang. 2018. A co-attention neural network model for emotion cause analysis with emotional context awareness. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 4752–4757, Brussels, Belgium. Association for Computational Linguistics. find
  35. Saif Mohammad, Xiaodan Zhu, and Joel Martin. 2014. Semantic role labeling of emotions in tweets. In Proceedings of the 5th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, pages 32–41, Baltimore, Maryland. Association for Computational Linguistics. find
  36. Klaus R. Scherer. 2005. What are emotions? And how can they be measured? Social Science Information, 44(4):695–729. ↗
  37. Shabnam Tafreshi and Mona Diab. 2018. Sentence and clause level emotion annotation, detection, and classification in a multi-genre corpus. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), pages 1246–1251, Miyazaki, Japan. European Language Resources Association (ELRA). find
  38. Alena Neviarouskaya and Masaki Aono. 2013. Extracting causes of emotions from text. In Proceedings of the Sixth International Joint Conference on Natural Language Processing, pages 932–936, Nagoya, Andrew J. Viterbi. 1967. Error bounds for convolutional codes and an asymptotically optimum decoding algorithm. IEEE Transactions on Information Theory, 13(2):260–269. find
  39. Penghui Wei, Jiahao Zhao, and Wenji Mao. 2020. Effective inter-clause modeling for end-to-end emotion-cause pair extraction. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 3171–3181, Online. Association for Computational Linguistics. find
  40. Rui Xia and Zixiang Ding. 2019. Emotion-cause pair extraction: A new task to emotion analysis in texts. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1003–1012, Florence, Italy. Association for Computational Linguistics. find
  41. Rui Xia, Mengran Zhang, and Zixiang Ding. 2019. RTHN: A RNN-Transformer Hierarchical Network for Emotion Cause Extraction. In Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence (IJCAI-19), pages 5285–5291, Macao, China. International Joint Conferences on Artificial Intelligence. find
  42. Bo Xu, Hongfei Lin, Yuan Lin, Yufeng Diao, Lian Yang, and Kan Xu. 2019. Extracting emotion causes using learning to rank methods from an information retrieval perspective. IEEE Access, 7:15573–15583. ↗
  43. Ruifeng Xu, Jiannan Hu, Qin Lu, Dongyin Wu, and Lin Gui. 2017. An ensemble approach for emotion cause detection with event extraction and multi-kernel svms. Tsinghua Science and Technology, 22(6):646–659. ↗
  44. Shuntaro Yada, Kazushi Ikeda, Keiichiro Hoashi, and Kyo Kageura. 2017. A bootstrap method for automatic rule acquisition on emotion cause extraction. In 2017 IEEE International Conference on Data Mining Workshops (ICDMW), pages 414–421. find
  45. Bishan Yang and Claire Cardie. 2013. Joint inference for fine-grained opinion extraction. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1640–1649, Sofia, Bulgaria. Association for Computational Linguistics.
  46. Xinyi Yu, Wenge Rong, Zhuo Zhang, Yuanxin Ouyang, and Zhang Xiong. 2019. Multiple level hierarchical network-based clause selection for emotion cause extraction. IEEE Access, 7:9071–9079. *Figure 7: Results of the three different models across the five different datasets* ↗

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/oberlander-2020-token-sequence-labeling-vs-clause/