Elamur Research⌕
☰
home / papers / 2010 / Resolving Object and Attribute Coreference in Opinion Mining

Resolving Object and Attribute Coreference in Opinion Mining

Summary

Opinion mining often fails to determine what a pronoun or other referring phrase points to, such as "it" for a product or "they" for one of its parts. The opinion then loses its target or gets attached to the wrong one. Ding and Liu define a task of linking mentions of objects (products) and their attributes in opinionated text. They treat it as supervised pairwise classification and extend the Soon et al. (2001) decision-tree framework with two opinion-aware features:

  • Sentiment consistency (SC): writers tend to keep the same polarity toward an entity across normal and comparative sentences.
  • Entity–opinion association (EOA): a PMI-based score, mined from the corpus, of how plausibly an opinion word describes a given object or attribute. For example, "expensive" suits a camera but not picture quality.

The authors annotated forum and blog posts in three domains. The full system clearly beats a Soon-style baseline and a centering-theory baseline. This shows that sentiment signals can guide reference resolution and help keep opinions attached to the correct targets.

Key points

  • Data: 453 posts and 3,939 sentences covering cellphones (168 posts), TVs (173) and cars (112). Mentions are annotated with OBJ/ATTR types and coreference links, and the authors plan to release the annotations.

  • Pipeline:

    • Stanford POS tagger and a CRF-based NP chunker for preprocessing.
    • Pairwise feature vectors that include string similarity and a keywords-between feature.
    • A J48 decision tree in WEKA, evaluated with 10-fold cross-validation.

  • Baselines:

    • Baseline1: Soon et al. (2001), reduced to 9 features.
    • Baseline2: semantic centering theory following Fang et al. (2009).

  • F-scores (full system vs. Baseline1 vs. Baseline2):

    DomainFull systemBaseline1Baseline2
    Cellphones0.720.610.67
    TVs0.730.640.68
    Cars0.750.660.73

  • Feature contributions: SC alone adds about 5–6 F points over Baseline1, and EOA adds a further 2–4. Paired t-tests show significance at 95% in most comparisons; the gain over Baseline2 on cars is significant only at 90%.

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: Coreference resolution

Cited by

9 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 8 not yet.

Cite

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

@inproceedings{ding2010resolving,
  title = {{Resolving Object and Attribute Coreference in Opinion Mining}},
  author = {Ding, Xiaowen and Liu, Bing},
  booktitle = {Proceedings of the 23rd International Conference on Computational Linguistics (Coling 2010)},
  year = {2010},
  url = {https://aclanthology.org/C10-1031/}
}

References

38 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. A. Andreevskaia and S. Bergler. 2006. Mining WordNet for Fuzzy Sentiment: Sentiment Tag Extraction from WordNet Glosses. EACL’06. find
  2. J. Carbonell and R. Brown. 1988. Anaphora resolution: a multi-strategy approach. COLING’1988. find our card
  3. G. Carenini, R. Ng, and A. Pauls. 2006. Interactive multimedia summaries of evaluative text. IUI’06. find
  4. X. Ding, B. Liu and L. Zhang. 2009. Entity Discovery and Assignment for Opinion Mining Application. KDD’09. find
  5. A. Esuli and F. Sebastiani. 2006. Determining Term Subjectivity and Term Orientation for Opinion Mining, EACL’06. find
  6. J. Fan, K. Barker and B. Porter. 2005. Indirect anaphora resolution as semantic path search. K-CAP’05. find
  7. C. Gasperin and T. Briscoe. 2008. Statistical anaphora resolution in biomedical texts. COLING'08 find our card
  8. B. J. Grosz, A. K. Joshi and S. Weinstein. 1995. Centering: a framework for modeling the local coherence of discourse. Computational Linguistics, 21(2). find our card
  9. V. Hatzivassiloglou and K. McKeown. 1997. Predicting the Semantic Orientation of Adjectives. ACL-EACL’97. find
  10. M. Hu and B. Liu. 2004. Mining and summarizing customer reviews. KDD’04. find
  11. N. Jindal, and B. Liu. 2006. Mining Comparative Sentences and Relations. AAAI’06. find
  12. H. Kanayama and T. Nasukawa. 2006. Fully Automatic Lexicon Expansion for Domain-Oriented Sentiment Analysis. EMNLP’06. find
  13. S. Kim and E. Hovy. 2004. Determining the Sentiment of Opinions. COLING’04. find
  14. N. Kobayashi, K. Inui and Y. Matsumoto. 2007. Extracting Aspect-Evaluation and Aspect-of Relations in Opinion Mining. EMNLP-CoNLL’07. find
  15. F. Kong, G. Zhou, Q. Zhu and P. Qian. 2009. Employing the Centering Theory in Pronoun Resolution from the Semantic Perspective. EMNLP’09. find our card
  16. L.-W. Ku, Y.-T. Liang and H.-H. Chen. 2006. Opinion Extraction, Summarization and Tracking in News and Blog Corpora. CAAW'06. find
  17. B. Liu. 2006. Web Data Mining, Springer. find
  18. R. McDonald, K. Hannan, T Neylon, M. Wells, and J.Reynar. 2007. Structured Models for Fine-to-Coarse Sentiment Analysis. ACL-07. find
  19. Q. Mei, X. Ling, M. Wondra, H. Su, and C. Zhai. 2007. Topic Sentiment Mixture: Modeling Facets and Opinions in Weblogs. WWW’07. find
  20. T. S. Morton. 2000. Coreference for NLP applications. ACL’00. find
  21. V. Ng and C. Cardie. 2002. Improving machine learning approaches to coreference resolution. ACL’02. find our card
  22. V. Ng. 2007. Semantic Class Induction and Coreference Resolution. ACL’07. find our card
  23. V. Ng. 2008. Unsupervised Models for Coreference Resolution. EMNLP’08. find our card
  24. N. Nicolov, F. Salvetti and S. Ivanova, Sentiment analysis: Does coreference matter? AISB'2008. find
  25. B. Pang, L. Lee, and S. Vaithyanathan. 2002. Thumbs up? Sentiment Classification Using Machine Learning Techniques. EMNLP’02. find
  26. H. Poon and P. Domingos. 2008. Joint Unsupervised Coreference Resolution with Markov Logic. EMNLP’08, 650–659. find our card
  27. A-M. Popescu and O. Etzioni. 2005. Extracting product features and opinions from reviews. EMNLP’05. find
  28. Qiu, Guang, B. Liu, J. Bu and C. Chen. 2009 Expanding Domain Sentiment Lexicon through Double Propagation. find
  29. W. M. Soon, H. T. Ng and D. Lim. 2001. A machine learning approach to coreference resolution of noun phrase. Computational Linguistics, 27(4). find
  30. V. Stoyanov, C. Cardie. 2006. Partially supervised coreference resolution for opinion summarization through structured rule learning. EMNLP’06. find our card
  31. I. Titov and R. McDonald. 2008, A joint model of text and aspect ratings for sentiment summarization, ACL’08. find
  32. J. Tetreault. 2001. A corpus-based evaluation of centering and pronoun resolution. Computational Linguistics. 27(4):507-520. find our card
  33. P. Turney. 2002. Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews. ACL’02. find
  34. Y. Versley, A. Moschitti, M. Poesio and X. Yang. 2008. Coreference systems based on kernels methods. COLING’08. find our card
  35. T. Wilson, J. Wiebe, and R. Hwa. 2004. Just how mad are you? Finding strong and weak opinion clauses. AAAI’04. find
  36. X. F. Yang㸪 J. Su and C. L. Tan. 2005. Improving Pronoun Resolution Using Statistics - Based Semantic Compatibility Information. ACL’05. find our card
  37. X. F. Yang and J. Su. 2007. Coreference Resolution Using Semantic Relatedness Information from Automatically Discovered Patterns. ACL’07. find our card
  38. G. D. Zhou and J. Su. 2004. A high-performance coreference resolution system using a multi-agent strategy. COLING’04. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/ding-2010-resolving-object-attribute-coreference-opinion/