Elamur Research⌕
☰
home / papers / 2013 / Analyses of the Association between Discourse Relation and Sentiment Polarity with a Chinese Human-Annotated Corpus

Analyses of the Association between Discourse Relation and Sentiment Polarity with a Chinese Human-Annotated Corpus

Written for the Research library (claude-opus-5-5) from the paper text, in our own words: the license "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)" does not allow redistributing the text itself. Original — aclanthology.org/W13-2309.pdf

Summary

The paper asks whether the type of an explicit discourse relation in Chinese predicts how sentiment is distributed across its two arguments. No public Chinese discourse corpus with sentiment labels existed, so the authors built one. They used a dictionary of 808 Chinese discourse markers, mapped to the four top-level PDTB classes, to select sentences from the Chinese portion of ClueWeb09. Each selected sentence has two short clauses joined by exactly one marker. In total, 7,638 sentences were labeled for clause-level and sentence-level polarity and for the relation class.

The analysis shows that relation type and sentiment are clearly associated. Temporal and Expansion instances tend to be objective. Comparison, and to a lesser degree Contingency, carry most of the sentiment, especially negative sentiment and polarity shifts. The polarity of the whole sentence usually follows the second clause. The data also reflect a Chinese "good words ahead" habit, where praise comes before criticism. The resource matters for Chinese discourse parsing and discourse-aware sentiment analysis, and it is released via NTU Chinese Discourse Resources.

Key points

  • Source data: The base corpus is the Chinese part of ClueWeb09, with about 9.6 billion POS-tagged sentences in roughly 172 million documents.
  • Sampling criteria: Each sentence has exactly two clauses, each clause has at most 20 characters, and the sentence contains a single marker from the dictionary. Inter-sentential-only markers were excluded.
  • Annotation: 87 native speakers took part, with three annotators per instance and labels decided by majority vote. Fleiss' Kappa scores, all in the moderate range, were:
    • first-clause polarity: 0.49
    • second-clause polarity: 0.50
    • whole-sentence polarity: 0.47
    • relation class: 0.41
  • Comparison relations: Neutral-Neutral is the most frequent transition (23.14%), followed closely by Positive-Negative (22.71%). The To-Negative share is 42.30%, well above that of the other classes.
  • Marker behavior: Markers such as 卻 and 否則 tend to introduce a negative turn. Markers such as 雖然…但 and 終於 tend to lead to a positive outcome. The authors identify marker ambiguity as a direction for future work.

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

Topics: Discourse analysis · Treebanks

Cited by

4 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 3 not yet.

Cite

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

@inproceedings{huang2013analyses,
  title = {{Analyses of the Association between Discourse Relation and Sentiment Polarity with a Chinese Human-Annotated Corpus}},
  author = {Huang, Hen-Hsen and Yu, Chi-Hsin and Chang, Tai-Wei and Lin, Cong-Kai and Chen, Hsin-Hsi},
  booktitle = {Proceedings of the 7th Linguistic Annotation Workshop and Interoperability with Discourse},
  year = {2013},
  url = {https://aclanthology.org/W13-2309/}
}

References

21 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Lynn Carlson and Daniel Marcu. 2001. Discourse Tagging Reference Manual. http://www.isi.edu/~marcu/discourse/tagging-ref-manual.pdf ↗
  2. Lynn Carlson, Daniel Marcu, and Mary Ellen Okurowski. 2002. RST Discourse Treebank. Linguistic Data Consortium, Philadelphia. find
  3. Shou-Yi Cheng. 2006. Corpus-Based Coherence Relation Tagging in Chinese Discourse. Master’s Thesis, National Chiao Tung University, Hsinchu, Taiwan. find
  4. Xianghui Cheng and Xiaolin Tian. 1989. Xian dai Han yu ( 現代漢語 ), San lian shu dian (三聯書店), Hong Kong. find
  5. Ann Devitt and Khurshid Ahmad. 2007. Sentiment polarity identification in financial news: a cohesion-based approach. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics (ACL 2007), pages 984-991, Prague, Czech Republic. find
  6. Hugo Hernault, Helmut Prendinger, David A. duVerle, and Mitsuru Ishizuka. 2010. HILDA: A Discourse Parser Using Support Vector Machine Classification. Dialogue and Discourse, 1(3): 1-33. find our card
  7. Hen-Hsen Huang and Hsin-Hsi Chen. 2011. Chinese discourse relation recognition. In Proceedings of the 5th International Joint Conference on Natural Language Processing (IJCNLP 2011). pages 1442-1446, Chiang Mai, Thailand. find our card
  8. Hen-Hsen Huang and Hsin-Hsi Chen. 2012a. Contingency and comparison relation labeling and structure prediction in Chinese sentences. In Proceedings of the 13th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDI-AL 2012), pages 261-269, Seoul, South Korea. find
  9. Hen-Hsen Huang and Hsin-Hsi Chen. 2012b. An Annotation System for Development of Chinese Discourse Corpus. In Proceedings of the 24th International Conference on Computational Linguistics (COLING 2012): Demonstration Papers, pages 223-230, Mumbai, India. find
  10. Ben Hutchinson. 2004. Acquiring the meaning of discourse markers. In Proceedings of the 42nd Annual Meeting of the Association for Computational Linguistics (ACL 2004), pages 684-691, Barcelona, Spain. find our card
  11. Soo-Min Kim and Eduard Hovy. 2004. Determining the sentiment of opinions. In Proceedings of the 20th International Conference on Computational Linguistics (COLING-04), pages 1367-1373, Geneva, Switzerland. find
  12. Ziheng Lin, Min-Yen Kan, and Hwee Tou Ng. 2009. Recognizing implicit discourse relations in the Penn Discourse Treebank. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing (EMNLP 2009), pages 343-351. find our card
  13. Shuxiang Lu. 2007. Eight Hundred Words of The Contemporary Chinese (Xian dai Han yu Ba bai Ci), China Social Sciences Press. find
  14. Bo Pang and Lillian Lee. 2008. Opinion Mining and Sentiment Analysis. Foundations and Trends in Information Retrieval, 2(1-2): 1-135. find
  15. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Eleni Miltsakaki, Livio Robaldo, Aravind Joshi, and Bonnie Webber. 2007. The Penn Discourse Treebank 2.0 Annotation Manual. The PDTB Research Group. find
  16. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Eleni Miltsakaki, Livio Robaldo, Aravind Joshi, and Bonnie Webber. 2008. The Penn Discourse Tree-Bank 2.0. In Proceedings of the 6th Language Resources and Evaluation Conference (LREC 2008), pages 2961-2968, Marrakech, Morocco. find
  17. Radu Soricut and Daniel Marcu. 2003. Sentence level discourse parsing using syntactic and lexical information. In Proceedings of Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics (HLT/NAACL 2003), pages 149-156, Edmonton, Canada. find our card
  18. Florian Wolf and Edward Gibson. 2005. Representing Discourse Coherence: A Corpus-Based Analysis. Computational Linguistics, 31(2): 249-287. find our card
  19. Chi-Hsin Yu, Yi-jie Tang and Hsin-Hsi Chen. 2012. Development of a web-scale Chinese word N-gram corpus with parts of speech information. In Proceedings the 8th International Conference on Language Resources and Evaluation (LREC 2012), pages 320-324, Istanbul, Turkey. find
  20. Zhi-Min Zhou, Yu Xu, Zheng-Yu Niu, Man Lan, Jian Su, and Chew Lim Tan. 2010. Predicting discourse connectives for implicit discourse relation recognition. In Proceedings of the 23rd International Conference on Computational Linguistics (COLING 2010): Posters, pages 1507-1514. find our card
  21. Yuping Zhou and Nianwen Xue. 2012. PDTB-style discourse annotation of Chinese text. In Proceedings the 50th Annual Meeting of the Association for Computational Linguistics (ACL 2012), pages 69-77, Jeju, South Korea. find

↑ back to top