Elamur ResearchβŒ•
☰
home / papers / 2017 / Magnets for Sarcasm: Making Sarcasm Detection Timely, Contextual and Very Personal

Magnets for Sarcasm: Making Sarcasm Detection Timely, Contextual and Very Personal

Summary

Sarcasm on Twitter often cannot be recognized from the tweet text alone, because the reader needs context and some sense of the author's attitude. Ghosh and Veale examine three questions: whether a speaker's mood helps detection, how the conversational context can be added to a neural model, and how the model can keep pace with shifting public opinion. They estimate the author's current mood from the 50 tweets posted just before the target tweet, scored on the 11 AnalyzeWords/LIWC dimensions. For context, they use the tweet to which the target is a reply. The model combines CNN, LSTM and DNN components. Training labels come from a feedback scheme in which the authors of the tweets themselves confirm whether they were being sarcastic, rather than relying on outside annotators.

Adding either the author's mood or the context tweet gives similarly large gains over text alone, but using both together adds little beyond either one. Retraining each week during the 2016 US campaign further improved results. The released dataset, Sarcasm Magnet, pairs each tweet with its context and a snapshot of its author's psychological profile.

Key points

  • Mood features: Each author is profiled on 11 AnalyzeWords dimensions, scored 0–100, including Angry, Analytic, Upbeat and Arrogant. The profiles were analysed with KNN clustering (k=30) and Spearman correlations against sarcastic behaviour.
  • Sarcasm magnets: In a set of 18K sarcastic tweets, the most frequent targets were @hillaryClinton, @realDonaldTrump, @bernieSanders, @AP and @megynKelly. Tweets aimed at @realDonaldTrump had the highest mean anger, yet 63% of their affective words were positive.
  • Model inputs: Configurations combine the target tweet, with or without the addressee (TTIA/TTEA), the context tweet (CT) and the psychological dimensions (PD).
  • Incremental retraining: In August 2016, four weekly batches of feedback-annotated tweets were added. On a fixed test set of 1,200 tweets (550 sarcastic, 650 non-sarcastic), F-score rose from 0.752 to 0.85.
  • Labelling approach: Labels come directly from the authors, so the model learns what the writer intended rather than how an outside annotator interpreted the tweet.

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

Fields: Linguistics & languages

Cited by

27 papers of our library cite it (of the 140673 whose bibliographies we read); 12 of them have a page here, 15 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/d17-1050

@inproceedings{ghosh2017magnets,
  title = {{Magnets for Sarcasm: Making Sarcasm Detection Timely, Contextual and Very Personal}},
  author = {Ghosh, Aniruddha and Veale, Tony},
  booktitle = {Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing},
  year = {2017},
  doi = {10.18653/v1/d17-1050},
  url = {https://doi.org/10.18653/v1/d17-1050}
}

References

36 references, 12 of them with a card of ours. β†— leads to the work, Β«findΒ» searches for one we have no id for yet.

  1. Deanna W. Gibbs and Herbert H. Clark. 1992. Coordinating beliefs in conversation. Journal of Memory and Language 31(2):183–194. doi:10.1016/0749-596X(92)90010-U. β†—
  2. Silvio Amir, Byron C Wallace, Hao Lyu, and Paula Carvalho Mário J Silva. 2016. Modelling context with user embeddings for sarcasm detection in social media. arXiv preprint arXiv:1607.00976 . β†— our card
  3. Raymond W. Gibbs and Herbert L. Colston. 2007. Irony in language and thought: A cognitive science reader. Psychology Press. https://books.google.ie/books?isbn=0805860622. β†—
  4. David Bamman and Noah A Smith. 2015. Contextualized sarcasm detection on twitter. In Ninth International AAAI Conference on Web and Social Media. http://homes.cs.washington.edu/ na-smith/papers/bamman+smith.icwsm15.pdf. β†—
  5. Roberto González-Ibánez, Smaranda Muresan, and Nina Wacholder. 2011. Identifying sarcasm in twitter: a closer look. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies: short papers-Volume 2. Association for Computational Linguistics, pages 581–586. http://www.aclweb.org/anthology/P11-2102. β†— our card
  6. Penelope Brown and Stephen C Levinson. 1978. Universals in language usage: Politeness phenomena. Cambridge University Press. http://www.mpi.nl/publications/escidoc-66660/. β†—
  7. John D Campbell and Albert N Katz. 2012. Are there necessary conditions for inducing a sense of sarcastic irony? Discourse Processes 49(6):459–480. http://dx.doi.org/10.1080/0163853X.2012.687863. β†—
  8. Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long short-term memory. Neural computation 9(8):1735–1780. http://dl.acm.org/citation.cfm?id=1246450. β†—
  9. Carol A Capelli, Noreen Nakagawa, and Cary M Madden. 1990. How children understand sarcasm: The role of context and intonation. Child Development 61(6):1824–1841. https://www.jstor.org/stable/1130840. β†—
  10. Aditya Joshi, Vinita Sharma, and Pushpak Bhattacharyya. 2015. Harnessing context incongruity for sarcasm detection. In Association for Computational Linguistics Volume 2: Short Papers. pages 757–762. https://www.aclweb.org/anthology/P/P15/P15-2124.pdf. β†— our card
  11. Fabio Celli, Arindam Ghosh, Firoj Alam, and Giuseppe Riccardi. 2016. In the mood for sharing contents: Emotions, personality and interaction styles in the diffusion of news. Information Processing & Management 52(1):93–98. find
  12. Anupam Khattri, Aditya Joshi, Pushpak Bhattacharyya, and Mark Carman. 2015. Your sentiment precedes you: Using an author’s historical tweets to predict sarcasm. In Proceedings of the 6th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis. Association for Computational Linguistics, Lisboa, Portugal, pages 25–30. http://aclweb.org/anthology/W15-2905. β†— our card
  13. Mathieu Cliche. 2014. The sarcasm detector. http://www.thesarcasmdetector.com/. β†—
  14. Paul T Costa and Robert R McCrae. 2008. The revised neo personality inventory (neo-pi-r). The SAGE handbook of personality theory and assessment 2:179–198. β†—
  15. Roger J Kreuz. 1996. The use of verbal irony: Cues and constraints. Metaphor: Implications and applications pages 23–38. find
  16. Dmitry Davidov, Oren Tsur, and Ari Rappoport. 2010. Semi-supervised recognition of sarcasm in twitter and amazon. In Proceedings of the Fourteenth Conference on Computational Natural Language Learning. Association for Computational Linguistics, Uppsala, Sweden, pages 107– 116. http://www.aclweb.org/anthology/W10-2914. β†— our card
  17. Roger J Kreuz and Sam Glucksberg. 1989. How to be sarcastic: The echoic reminder theory of verbal irony. Journal of Experimental Psychology: General 118(4):374. http://dx.doi.org/10.1037/0096-3445.118.4.374. β†—
  18. Shelly Dews and Ellen Winner. 1995. Muting the meaning a social function of irony. Metaphor and Symbol 10(1):3–19. http://dx.doi.org/10.1207/s15327868ms1001 2. β†—
  19. Roger J Kreuz and Kristen E Link. 2002. Asymmetries in the use of verbal irony. Journal of Language and Social Psychology 21(2):127–143. http://dx.doi.org/10.1177/02627X02021002002. β†—
  20. Shelly Dews and Ellen Winner. 1999. Obligatory processing of literal and nonliteral meanings in verbal irony. Journal of pragmatics 31(12):1579–1599. http://psycnet.apa.org/psycinfo/1999-01341-003. β†—
  21. Christine Liebrecht, Florian Kunneman, and Antal Van den Bosch. 2013. The perfect solution for detecting sarcasm in tweets #not. In Proceedings of the 4th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis. Association for Computational Linguistics, Atlanta, Georgia, pages 29–37. http://www.aclweb.org/anthology/W13-1605. β†— our card
  22. Aniruddha Ghosh and Tony Veale. 2016. Fracking sarcasm using neural network. In Proceedings of NAACL-HLT. pages 161–169. nition of sarcastic sentences in online product reviews. In Proceedings of the Fourth International Conference on Weblogs and Social Media. β†— our card
  23. Stephanie Lukin and Marilyn Walker. 2013. Really? well. apparently bootstrapping improves the performance of sarcasm and nastiness classifiers for online dialogue. In Proceedings of the Workshop on Language Analysis in Social Media. Association for Computational Linguistics, Atlanta, Georgia, pages 30–40. http://www.aclweb.org/anthology/W13-1104. β†— our card
  24. A. Utsumi. 2000a. Verbal irony as implicit display of ironic environment: Distinguishing ironic utterances from nonirony. Journal of Pragmatics http://www.utm.se.uec.ac.jp/ utsumi/paper/jop2000-utsumi.pdf. β†—
  25. Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg S Corrado, and Jeff Dean. 2013. Distributed representations of words and phrases and their compositionality. In Advances in neural information processing systems. pages 3111–3119. https://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf. β†—
  26. Akira Utsumi. 2000b. Verbal irony as implicit display of ironic environment: Distinguishing ironic utterances from nonirony. Journal of Pragmatics 32(12):1777–1806. β†—
  27. Jennifer Woodland and Daniel Voyer. 2011. Context and intonation in the perception of sarcasm. Metaphor and Symbol 26(3):227–239. http://dx.doi.org/10.1080/10926488.2011.583197. β†—
  28. James M. Olver and Todd A. Mooradian. 2003. Personality traits and personal values: a conceptual and empirical integration. Personality and Individual Differences 35(1):109 – 125. https://doi.org/http://dx.doi.org/10.1016/S0191-8869(02)00145-9. β†—
  29. Meishan Zhang, Yue Zhang, and Guohong Fu. 2016. Tweet sarcasm detection using deep neural network. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers. find our card
  30. The COL-ING 2016 Organizing Committee, pages 2449– 2460. http://aclweb.org/anthology/C16-1231. β†— our card
  31. Tomáš Ptáček, Ivan Habernal, and Jun Hong. 2014. Sarcasm detection on czech and english twitter. In Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: Technical Papers. Dublin City University and Association for Computational Linguistics, Dublin, Ireland, pages 213–223. http://www.aclweb.org/anthology/C14-1022. β†— our card
  32. Ashwin Rajadesingan, Reza Zafarani, and Huan Liu. 2015. Sarcasm detection on twitter: A behavioral modeling approach. In Proceedings of the Eighth ACM International Conference on Web Search and Data Mining. ACM, pages 97–106. http://dl.acm.org/citation.cfm?id=2685316. β†—
  33. Ellen Riloff, Ashequl Qadir, Prafulla Surve, Lalindra De Silva, Nathan Gilbert, and Ruihong Huang. 2013. Sarcasm as contrast between a positive sentiment and negative situation. In Empirical Methods on Natural Language Processing. volume 13, pages 704–714. http://www.anthology.aclweb.org/D/D13/D13-1066.pdf. β†— our card
  34. SG Shamay-Tsoory, Rachel Tomer, and Judith Aharon-Peretz. 2005. The neuroanatomical basis of understanding sarcasm and its relationship to social cognition. Neuropsychology 19(3):288. http://www.apa.org/pubs/journals/releases/neu-193288.pdf. β†—
  35. Yla R Tausczik and James W Pennebaker. 2010. The psychological meaning of words: Liwc and computerized text analysis methods. Journal of language and social psychology 29(1):24–54. http://dx.doi.org/10.1177/0261927X09351676. β†—
  36. Oren Tsur, Dmitry Davidov, and Ari Rappoport. 2010. Icwsm-a great catchy name: Semi-supervised recog- find

↑ back to top

This page in Markdown β†’ https://md.research.elamur.ai/free/papers/ghosh-2017-magnets-sarcasm-making-sarcasm-detection/