Elamur Research⌕
☰
home / papers / 2017 / Overcoming Language Variation in Sentiment Analysis with Social Attention

Overcoming Language Variation in Sentiment Analysis with Social Attention

Paper · 2017 · Yi Yang and Jacob Eisenstein

Summary

Social media language differs from author to author. A word such as "sick" can express disgust for one writer and enthusiasm for another, which misleads sentiment classifiers trained on pooled data. Yang and Eisenstein address this with linguistic homophily: people who are socially connected tend to use words in similar ways.

A pilot study on SemEval Twitter data supports the idea. Errors of a simple lexicon-based classifier cluster more on real social graphs than on randomly rewired copies of those graphs.

The authors then propose SOCIAL ATTENTION, a neural mixture-of-experts model. Several convolutional basis classifiers are combined using author-specific weights, which are derived from network embeddings of each user. As a result, nearby authors receive similar classifiers. This gives personalization even for users with no labeled messages and no demographic metadata.

On the SemEval Twitter test sets and on Ciao product reviews, the model beats comparable architectures without social information. It also exceeds previously published SemEval results. The method offers a general, network-based way to make NLP robust to social variation.

Key points

  • Twitter data: Training and tuning used the SemEval Twitter 2013 train and dev sets, with testing on the 2013, 2014 and 2015 test sets. The authors built three undirected author graphs (FOLLOWER, MENTION, RETWEET) for 7,438 training-set authors, crawling up to 3,200 recent tweets per user via the Twitter API.
  • Homophily pilot: The lexicon-based classifier used the Tang et al. (2014) lexicons (1,474 positive and 1,956 negative words). Its errors were compared against degree-preserving rewired networks (networkx double edge swap). Observed assortativity was higher in nearly every case, and no single network type was clearly best.
  • Model design: Each author's prediction is a weighted mixture of K CNN basis classifiers. The weights come from an attention component over author embeddings, learned with a large-scale network embedding method (Tang et al., 2015b).
  • Second domain: The model was also applied to Ciao product reviews, where author embeddings were trained on trust relationships between reviewers.
  • Results: SOCIAL ATTENTION improves accuracy by roughly 2–3% over neural and ensemble baselines with the social information removed. On the SemEval Twitter test sets, it also surpasses every result reported in earlier published 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.

 Institutions

🎓Georgia Institute of TechnologyAtlanta · United States133 works · 24,497 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 1Research subjects and languages the work is about.Sentiment & emotion · 972 works

Cite

Cite as Yi Yang and Jacob Eisenstein. 2017. Overcoming Language Variation in Sentiment Analysis with Social Attention. Transactions of the Association for Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.1162/tacl_a_00062

@article{yang2017overcoming,
  title = {{Overcoming Language Variation in Sentiment Analysis with Social Attention}},
  author = {Yi Yang and Jacob Eisenstein},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2017},
  doi = {10.1162/tacl\_a\_00062},
  url = {https://doi.org/10.1162/tacl_a_00062}
}

References

54 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Mohammad Al Boni, Keira Qi Zhou, Hongning Wang, and Matthew S Gerber. 2015. Model adaptation for personalized opinion analysis. In Proceedings of the Association for Computational Linguistics (ACL). Ramon F Astudillo, Silvio Amir, Wang Ling, Mário Silva, and Isabel Trancoso. 2015. Learning word rep… find
  2. AiTi Aw, Min Zhang, Juan Xiao, and Jian Su. 2006. A phrase-based statistical model for SMS text normalization. In Proceedings of the Association for Computational Linguistics (ACL).
  3. Justin Basilico and Thomas Hofmann. 2004. Unifying collaborative and content-based filtering. In Proceedings of the International Conference on Machine Learning (ICML). find
  4. John Blitzer, Ryan McDonald, and Fernando Pereira. 2006. Domain adaptation with structural correspondence learning. In Proceedings of Empirical Methods for Natural Language Processing (EMNLP).
  5. Su Lin Blodgett, Lisa Green, and Brendan O’Connor. 2016. Demographic dialectal variation in social media: A case study of african-american english. In Proceedings of Empirical Methods for Natural Language Processing (EMNLP).
  6. Antoine Bordes, Nicolas Usunier, Alberto Garcia-Duran, Jason Weston, and Oksana Yakhnenko. 2014. Translating embeddings for modeling multi-relational data. In Neural Information Processing Systems (NIPS).
  7. John S Breese, David Heckerman, and Carl Kadie. 1998. Empirical analysis of predictive algorithms for collaborative filtering. In Proceedings of Uncertainty in Artificial Intelligence (UAI). find
  8. John Bryden, Sebastian Funk, and Vincent Jansen. 2013. Word usage mirrors community structure in the online social network twitter. EPJ Data Science, 2(1). find
  9. Ciprian Chelba and Alex Acero. 2006. Adaptation of maximum entropy capitalizer: Little data can help a lot. Computer Speech & Language, 20(4). Hal Daumé III. 2007. Frustratingly easy domain adaptation. In Proceedings of the Association for Computational Linguistics (ACL).
  10. Penelope Eckert and Sally McConnell-Ginet. 2003. find
  11. Language and Gender. Cambridge University Press. Jacob Eisenstein, Brendan O’Connor, Noah A. Smith, and Eric P. Xing. 2010. A latent variable model for geographic lexical variation. In Proceedings of Empirical Methods for Natural Language Processing (EMNLP).
  12. Jacob Eisenstein. 2013. What to do about bad language on the internet. In Proceedings of the North American Chapter of the Association for Computational Linguistics (NAACL).
  13. Jacob Eisenstein. 2015. Systematic patterning in phonologically-motivated orthographic variation. Journal of Sociolinguistics, 19. find
  14. Marcello Federico. 1996. Bayesian estimation methods for n-gram language model adaptation. In Proceedings of International Conference on Spoken Language (ICSLP). find
  15. Jenny R. Finkel and Christopher Manning. 2009. Hierarchical bayesian domain adaptation. In Proceedings of the North American Chapter of the Association for Computational Linguistics (NAACL). find
  16. John L Fischer. 1958. Social influences on the choice of a linguistic variant. Word, 14. find
  17. Santo Fortunato. 2010. Community detection in graphs. Physics Reports, 486(3). find
  18. African American English: A Linguistic Introduction. Cambridge University Press. Aric A. Hagberg, Daniel A Schult, and P Swart. 2008. Exploring network structure, dynamics, and function using networkx. In Proceedings of the 7th Python in Science Conferences (SciPy). Matthias Hagen, Martin Potthast,… find
  19. Hussam Hamdan, Patrice Bellot, and Frederic Bechet. 2015. lsislif: Feature extraction and label weighting for sentiment analysis in twitter. In Proceedings of the 9th International Workshop on Semantic Evaluation. find
  20. Dirk Hovy. 2015. Demographic factors improve classification performance. In Proceedings of the Association for Computational Linguistics (ACL).
  21. Xia Hu, Lei Tang, Jiliang Tang, and Huan Liu. 2013. Exploiting social relations for sentiment analysis in microblogging. In Proceedings of Web Search and Data Mining (WSDM). find
  22. Bernardo Huberman, Daniel M. Romero, and Fang Wu. 2008. Social networks that matter: Twitter under the microscope. First Monday, 14(1). find
  23. Robert A Jacobs, Michael I Jordan, Steven J Nowlan, and Geoffrey E Hinton. 1991. Adaptive mixtures of local experts. Neural computation, 3(1). find
  24. Jing Jiang and ChengXiang Zhai. 2007. Instance weighting for domain adaptation in NLP. In Proceedings of the Association for Computational Linguistics (ACL).
  25. Nal Kalchbrenner, Edward Grefenstette, and Phil Blunsom. 2014. A convolutional neural network for modelling sentences. In Proceedings of the Association for Computational Linguistics (ACL).
  26. Yoon Kim. 2014. Convolutional neural networks for sentence classification. In Proceedings of Empirical Methods for Natural Language Processing (EMNLP).
  27. Diederik Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
  28. William Labov. 1963. The social motivation of a sound change. Word, 19(3). find
  29. Yann LeCun, Bernhard Boser, John S Denker, Donnie Henderson, Richard E Howard, Wayne Hubbard, and Lawrence D Jackel. 1989. Backpropagation applied to handwritten zip code recognition. Neural computation, 1(4). find
  30. Christopher J Leggetter and Philip C Woodland. 1995. Maximum likelihood linear regression for speaker adaptation of continuous density hidden markov models. Computer Speech & Language, 9(2). find
  31. Wang Ling, Chris Dyer, Alan Black, and Isabel Trancoso. 2015. Two/too simple adaptations of word2vec for syntax problems. In Proceedings of the North American Chapter of the Association for Computational Linguistics (NAACL).
  32. Miller McPherson, Lynn Smith-Lovin, and James M Cook. 2001. Birds of a feather: Homophily in social networks. Annual review of sociology. find
  33. Preslav Nakov, Zornitsa Kozareva, Alan Ritter, Sara Rosenthal, Veselin Stoyanov, and Theresa Wilson. 2013. Semeval-2013 task 2: Sentiment analysis in twitter. In Proceedings of the 7th International Workshop on Semantic Evaluation.
  34. Mark EJ Newman. 2003. The structure and function of complex networks. SIAM review, 45(2). find
  35. Sara Rosenthal and Kathleen McKeown. 2011. Age prediction in blogs: A study of style, content, and online behavior in pre- and Post-Social media generations. In Proceedings of the Association for Computational Linguistics (ACL). find
  36. Sara Rosenthal, Preslav Nakov, Svetlana Kiritchenko, Saif M Mohammad, Alan Ritter, and Veselin Stoyanov. 2015. Semeval-2015 task 10: Sentiment analysis in twitter. In Proceedings of the 9th International Workshop on Semantic Evaluation.
  37. Aliaksei Severyn and Alessandro Moschitti. 2015. Unitn: Training deep convolutional neural network for twitter sentiment classification. In Proceedings of the 9th International Workshop on Semantic Evaluation. find
  38. Xuehua Shen, Bin Tan, and ChengXiang Zhai. 2005. Implicit user modeling for personalized search. In Proceedings of the International Conference on Information and Knowledge Management (CIKM). find
  39. Kaisong Song, Shi Feng, Wei Gao, Daling Wang, Ge Yu, and Kam-Fai Wong. 2015. Personalized sentiment classification based on latent individuality of microblog users. In Proceedings of the 24th International Joint Conference on Artificial Intelligence (IJ-CAI). find
  40. Michael Speriosu, Nikita Sudan, Sid Upadhyay, and Jason Baldridge. 2011. Twitter polarity classification with label propagation over lexical links and the follower graph. In Proceedings of Empirical Methods for Natural Language Processing (EMNLP). find
  41. R. Sproat, A.W. Black, S. Chen, S. Kumar, M. Ostendorf, and C. Richards. 2001. Normalization of non-standard words. Computer Speech & Language, 15(3). find
  42. Chenhao Tan, Lillian Lee, Jie Tang, Long Jiang, Ming Zhou, and Ping Li. 2011. User-level sentiment analysis incorporating social networks. In Proceedings of Knowledge Discovery and Data Mining (KDD). find
  43. Jiliang Tang, Huiji Gao, and Huan Liu. 2012. mtrust: discerning multi-faceted trust in a connected world. In Proceedings of Web Search and Data Mining (WSDM). find
  44. Duyu Tang, Furu Wei, Bing Qin, Ming Zhou, and Ting Liu. 2014. Building large-scale twitter-specific sentiment lexicon: A representation learning approach. In Proceedings of the International Conference on Computational Linguistics (COLING). find
  45. Duyu Tang, Bing Qin, and Ting Liu. 2015a. Learning semantic representations of users and products for document level sentiment classification. In Proceedings of the Association for Computational Linguistics (ACL). find
  46. Jian Tang, Meng Qu, Mingzhe Wang, Ming Zhang, Jun Yan, and Qiaozhu Mei. 2015b. Line: Large-scale information network embedding. In Proceedings of the Conference on World-Wide Web (WWW). find
  47. Mike Thelwall. 2009. Homophily in MySpace. Journal of the American Society for Information Science and Technology, 60(2). find
  48. Matt Thomas, Bo Pang, and Lillian Lee. 2006. Get out the vote: Determining support or opposition from Congressional floor-debate transcripts. In Proceedings of Empirical Methods for Natural Language Processing (EMNLP).
  49. Peter Trudgill. 1974. Linguistic change and diffusion: Description and explanation in sociolinguistic dialect geography. Language in Society, 3(2). find
  50. Joseph B. Walther and Kyle P. D’Addario. 2001. The impacts of emoticons on message interpretation in computer-mediated communication. Social Science Computer Review, 19(3). find
  51. Zhen Wang, Jianwen Zhang, Jianlin Feng, and Zheng Chen. 2014. Knowledge graph embedding by translating on hyperplanes. In Proceedings of the National Conference on Artificial Intelligence (AAAI). find
  52. Robert West, Hristo Paskov, Jure Leskovec, and Christopher Potts. 2014. Exploiting social network structure for person-to-person sentiment analysis. Transactions of the Association for Computational Linguistics, 2. find
  53. Caroline Winterer. 2012. Where is America in the Republic of Letters? Modern Intellectual History, 9(03). find
  54. Fangzhao Wu and Yongfeng Huang. 2015. Personalized microblog sentiment classification via multi-task learning. In Proceedings of the National Conference on Artificial Intelligence (AAAI). find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/yang-2017-overcoming-language-variation-sentiment-analysis/