---
title: "LLT-PolyU: Identifying Sentiment Intensity in Ironic Tweets (2015)"
description: "The paper presents the LLT-PolyU entry to SemEval-2015 Task 11, which asks systems to predict a… 37 references, cited by 3 papers here."
authors: ["Hongzhi Xu", "Enrico Santus", "Anna Laszlo", "Chu-Ren Huang"]
year: 2015
venue: "Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015)"
source: "https://aclanthology.org/S15-2113/"
doi: "10.18653/v1/s15-2113"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/xu-2015-llt-polyu-identifying-sentiment-intensity/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/xu-2015-llt-polyu-identifying-sentiment-intensity/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2015](https://md.research.elamur.ai/free/en/papers/2015/page/3/) / LLT-PolyU: Identifying Sentiment Intensity in Ironic Tweets

# LLT-PolyU: Identifying Sentiment Intensity in Ironic Tweets

**Authors:** [Hongzhi Xu](https://md.research.elamur.ai/free/en/authors/hongzhi-xu/), [Enrico Santus](https://md.research.elamur.ai/free/en/authors/enrico-santus/), Anna Laszlo and [Chu-Ren Huang](https://md.research.elamur.ai/free/en/authors/chu-ren-huang/) · **Year:** 2015 · **Venue:** [Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015)](https://md.research.elamur.ai/free/en/venues/semeval/) ([SemEval 2015](https://md.research.elamur.ai/free/en/venues/semeval/2015/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/S15-2113/) · [PDF](https://aclanthology.org/S15-2113.pdf) · [DOI 10.18653/v1/s15-2113](https://doi.org/10.18653/v1/s15-2113) · [OpenAlex](https://openalex.org/W2101449828)

**Citations:** 16 citations (OpenAlex, 2026-10-14) · [3](https://md.research.elamur.ai/free/en/papers/xu-2015-llt-polyu-identifying-sentiment-intensity/cited-by/) in our library, of the 141,840 papers whose bibliographies we read

## Summary
The paper presents the LLT-PolyU entry to SemEval-2015 Task 11, which asks systems to predict a fine-grained sentiment intensity score for tweets that use figurative language: irony, sarcasm and metaphor. Ironic language can flip the polarity of a tweet's surface wording, so plain lexicon-based scoring often fails on it.

The authors first clean the tweets with heuristic rules. They collapse elongated vowels and capitals, reduce runs of ?/! marks, look up emoticons, and split glued out-of-vocabulary tokens. They then parse the tweets with the Stanford parser. On top of this they extract several feature groups:
- n-grams and dependency pairs;
- lexicon-based polarity sums with negation handling;
- polarity-shift features, designed for irony, that detect a sentiment verb attached to an opposite-polarity clause.

Features are filtered by frequency and by Pearson correlation with the target. A Weka RepTree regressor beat LibSVM SVR in cross-validation. The system ranked first on irony and third overall. This shows that explicit modelling of polarity clashes helps with ironic text.

## Key points
- **Data and setup:** the model is trained on 8,000 tweets and tested on 4,000. There were fifteen participating systems, and evaluation used cosine similarity and MSE.
- **Lexicons:** four sentiment lexicons are used: Opinion Lexicon, Afinn, MPQA and SentiWordnet. Their union and their intersection serve as two extra dictionaries. Negation is checked either in a 5-token window or along the dependency structure.
- **Feature filtering:** only features seen at least 3 times are kept. Values are scaled to [-1, 1], and a Pearson r threshold of 0.035 is applied, chosen from a sweep over r values.
- **Ablation (5-fold CV):**
  - MSE fell from 1.9847 for the mean-prediction baseline to 1.6393 with all features. Cosine rose from 0.8184 to 0.8543.
  - BiToken was the only feature group that hurt, so it was dropped from the final system.
- **Test results:**
  - Irony: cosine 0.918, MSE 0.671.
  - Sarcasm: cosine 0.896.
  - Metaphor: cosine 0.535.
  - Other: cosine 0.290.
  - Overall: cosine 0.687, MSE 2.602.

## Interesting ideas
- Elongated or capitalized words are normalized so the parser sees clean forms, but the emphasis type is kept as a separate tagged feature, so the signal is not lost.
- Glued out-of-vocabulary strings like "yeahright" are split by greedy maximal matching against a dictionary before parsing.
- Irony is spotted as a polarity clash: a positive sentiment verb governs a clause describing something negative, or the reverse.
- The same negation and polarity-shift checks are run both on a flat word window and on the dependency tree, and both variants are kept because each adds a little.
- The authors grow the system one feature group at a time against a trivial mean-prediction baseline and drop any group that does not help, which caught the harmful bigrams.

## For Mova

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

- **I1** (tokens): the tokenizer could emit the normalized lemma plus an "emphasis" attribute (elongation, caps, heavy punctuation), keeping parsing clean while preserving intensity cues for later stages.
- **I2** (tokens): a dictionary-driven splitter for OOV tokens in informal text would give the UD parser real words instead of unknown blobs, and each split would stay explainable.
- **I3** (frames): the irony component in prag could compare the evaluative polarity of the experiencer verb with the polarity of its complement frame, flagging a mismatch as an explicit, traceable irony cue.
- **I4** (tree): scoping negation through the UD tree rather than by linear distance can be validated against a window baseline, giving a measurable case for tree-based scope.
- **I5** (learn): evaluation could routinely include a trivial baseline as a negative control and greedy group-wise ablation, so that components that add no value are exposed.

*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

- [Hong Kong Polytechnic University](https://md.research.elamur.ai/free/en/universities/hong-kong-polytechnic-university/) — University, Hong Kong, China · 34 works · 4,476 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,086 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,889 works · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 798 works · [💭 Experimental and Cognitive Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/experimental-and-cognitive-psychology/) · 463 works

🧭 Topics (1): [😊 Sentiment & emotion](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) · 608 works

## Cite

Cite as: Hongzhi Xu, Enrico Santus, Anna Laszlo and Chu-Ren Huang. 2015. LLT-PolyU: Identifying Sentiment Intensity in Ironic Tweets. Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015).

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

```bibtex
@inproceedings{xu2015llt,
  title = {{LLT-PolyU: Identifying Sentiment Intensity in Ironic Tweets}},
  author = {Hongzhi Xu, Enrico Santus, Anna Laszlo and Chu-Ren Huang},
  booktitle = {Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015)},
  year = {2015},
  doi = {10.18653/v1/s15-2113},
  url = {https://doi.org/10.18653/v1/s15-2113}
}
```

- [← previous Chinese Event Coreference Resolution: An Unsupervised Probabilistic Model Rivaling Supervised Resolvers](https://md.research.elamur.ai/free/en/papers/chen-2015-chinese-event-coreference-resolution-unsupervised/)
- [next → The DCU Discourse Parser for Connective, Argument Identification and Explicit Sense Classification](https://md.research.elamur.ai/free/en/papers/wang-2015-dcu-discourse-parser-connective-argument/)
- [↑ back to papers of 2015](https://md.research.elamur.ai/free/en/papers/2015/page/3/)

## References

37 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Ahmed Abbasi, Hsinchun Chen, and Arab Salem. 2008. Sentiment analysis in multiple languages: Feature selection for opinion classification in web forums. ACM Transactions on Information Systems (TOIS), 26(3):12:1–12:34. [find](https://scholar.google.com/scholar?q=Ahmed+Abbasi%2C+Hsinchun+Chen%2C+and+Arab+Salem%2E+2008%2E+Sentiment+analysis+in+multiple+languages%3A+Feature+selection+for+opinion+classification+in+web+forums%2E+ACM+Transactions+on+Information+Systems+%28TOIS%29%2C)
2. Stefano Baccianella, Andrea Esuli, and Fabrizio Sebastiani. 2010. [Sentiwordnet 3.0: An enhanced lexical resource for sentiment analysis and opinion mining](https://md.research.elamur.ai/free/en/papers/baccianella-2010-sentiwordnet-3-0-enhanced-lexical/). In LREC, volume 10, pages 2200–2204.
3. Francesco Barbieri and Horacio Saggion. 2014. [Modelling irony in twitter](https://md.research.elamur.ai/free/en/papers/barbieri-2014-modelling-irony-twitter/). In Proceedings of the Student Research Workshop at the 14th Conference of the European Chapter of the Association for Computational Linguistics, pages 56–64.
4. Francesco Barbieri, Francesco Ronzano, and Horacio Saggion. 2014. Italian irony detection in twitter: a first approach. In The First Italian Conference on Computational Linguistics CLiC-it 2014 & the Fourth International Workshop EVALITA 2014, pages 28–32. [find](https://scholar.google.com/scholar?q=Francesco+Barbieri%2C+Francesco+Ronzano%2C+and+Horacio+Saggion%2E+2014%2E+Italian+irony+detection+in+twitter%3A+a+first+approach%2E+In+The+First+Italian+Conference+on+Computational+Linguistics+CLiC%2Dit+2014+%26+the+)
5. Cristina Bosco, Viviana Patti, and Andrea Bolioli. 2013. Developing corpora for sentiment analysis and opinion mining: the case of irony and senti-tut. IEEE Intelligent Systems, 28(2):55–63. [find](https://scholar.google.com/scholar?q=Cristina+Bosco%2C+Viviana+Patti%2C+and+Andrea+Bolioli%2E+2013%2E+Developing+corpora+for+sentiment+analysis+and+opinion+mining%3A+the+case+of+irony+and+senti%2Dtut%2E+IEEE+Intelligent+Systems%2C+28%282%29%3A55%E2%80%9363%2E)
6. Konstantin Buschmeier, Philipp Cimiano, and Roman Klinger. 2014. [An impact analysis of features in a classification approach to irony detection in product reviews](https://md.research.elamur.ai/free/en/papers/buschmeier-2014-impact-analysis-features-classification-approach/).
7. Paula Carvalho, Luı́s Sarmento, Mário J Silva, and Eugénio de Oliveira. 2009. Clues for detecting irony in user-generated contents: oh...!! it’s so easy;-). In Proceedings of the 1st international CIKM workshop on Topic-sentiment analysis for mass opinion, pages 53– 56. ACM. [find](https://scholar.google.com/scholar?q=Paula+Carvalho%2C+Lu%C4%B1%CC%81s+Sarmento%2C+Ma%CC%81rio+J+Silva%2C+and+Euge%CC%81nio+de+Oliveira%2E+2009%2E+Clues+for+detecting+irony+in+user%2Dgenerated+contents%3A+oh%2E%2E%2E%21%21+it%E2%80%99s+so+easy%3B%2D%29%2E+In+Proceedings+of+the+1st+international+C)
8. Chih-Chung Chang and Chih-Jen Lin. 2011. Libsvm: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2:1–27. [find](https://scholar.google.com/scholar?q=Chih%2DChung+Chang+and+Chih%2DJen+Lin%2E+2011%2E+Libsvm%3A+A+library+for+support+vector+machines%2E+ACM+Transactions+on+Intelligent+Systems+and+Technology%2C+2%3A1%E2%80%9327%2E)
9. Dmitry Davidov, Oren Tsur, and Ari Rappoport. 2010. [Semi-supervised recognition of sarcastic sentences in twitter and amazon](https://md.research.elamur.ai/free/en/papers/davidov-2010-semi-supervised-recognition-sarcasm-twitter/). In Proceedings of the Fourteenth Conference on Computational Natural Language Learning, pages 107–116.
10. Elena Filatova. 2012. [Irony and sarcasm: Corpus generation and analysis using crowdsourcing](https://md.research.elamur.ai/free/en/papers/filatova-2012-irony-sarcasm-corpus-generation-analysis/). In Proceedings of Language Resources and Evaluation Conference, pages 392–398.
11. Aniruddha Ghosh, Guofu Li, Tony Veale, Paolo Rosso, Ekaterina Shutova, Antonio Reyes, and John Barnden. 2015. [Semeval-2015 task 11: Sentiment analysis of figurative language in twitter](https://md.research.elamur.ai/free/en/papers/ghosh-2015-semeval-2015-task-11-sentiment/). In Proceedings of the International Workshop on Semantic Evaluation (SemEval-2015), Co-located with NAACL and *SEM, Denver, Colorado, US, June 4-5.
12. Andrea Gianti, Cristina Bosco, Viviana Patti, Andrea Bolioli, and Luigi Di Caro. 2012. Annotating irony in a novel italian corpus for sentiment analysis. In Proceedings of the 4th Workshop on Corpora for Research on Emotion Sentiment and Social Signals, pages 1–7. [find](https://scholar.google.com/scholar?q=Andrea+Gianti%2C+Cristina+Bosco%2C+Viviana+Patti%2C+Andrea+Bolioli%2C+and+Luigi+Di+Caro%2E+2012%2E+Annotating+irony+in+a+novel+italian+corpus+for+sentiment+analysis%2E+In+Proceedings+of+the+4th+Workshop+on+Corpora+)
13. Raymond W Gibbs and Herbert L Colston. 2007. Irony in language and thought: A cognitive science reader. Psychology Press. [↗](https://doi.org/10.4324/9781410616685)
14. Roberto González-Ibáñez, Smaranda Muresan, and Nina Wacholder. 2011. [Identifying sarcasm in twitter: a closer look](https://md.research.elamur.ai/free/en/papers/ibanez-2011-identifying-sarcasm-twitter-closer-look/). In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies: short papers, volume 2, pages 581–586.
15. H Paul Grice, 1975. Logic and conversation, volume 3 of Syntax and Semantics, pages 41–58. Academic Press, New York. [find](https://scholar.google.com/scholar?q=H+Paul+Grice%2C+1975%2E+Logic+and+conversation%2C+volume+3+of+Syntax+and+Semantics%2C+pages+41%E2%80%9358%2E+Academic+Press%2C+New+York%2E)
16. Mark Hall, Eibe Frank, Geoffrey Holmes, Bernhard Pfahringer, Peter Reutemann, and Ian H. Witten. 2009. The weka data mining software: An update. SIGKDD Explorations, 11. [↗](https://doi.org/10.1145/1656274.1656278)
17. Yanfen Hao and Tony Veale. 2010. An ironic fist in a velvet glove: Creative mis-representation in the construction of ironic similes. Minds and Machines, 20(4):635–650. [↗](https://doi.org/10.1007/s11023-010-9211-1)
18. Jennifer Hay. 2001. The pragmatics of humor support. International Journal of Humor Research, 14:1–27. [↗](https://doi.org/10.1515/humr.14.1.55)
19. Minqing Hu and Bing Liu. 2004. Mining and summarizing customer reviews. In Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining, pages 168–177. ACM. [find](https://scholar.google.com/scholar?q=Minqing+Hu+and+Bing+Liu%2E+2004%2E+Mining+and+summarizing+customer+reviews%2E+In+Proceedings+of+the+tenth+ACM+SIGKDD+international+conference+on+Knowledge+discovery+and+data+mining%2C+pages+168%E2%80%93177%2E+ACM%2E)
20. Dan Klein and Christopher D. Manning. 2003. [Accurate unlexicalized parsing](https://md.research.elamur.ai/free/en/papers/klein-2003-accurate-unlexicalized-parsing/). In Proceedings of the 41st Annual Meeting on Association for Computational Linguistics-Volume 1, pages 423–430.
21. Helga Kotthoff. 2003. Responding to irony in different contexts: On cognition in conversation. Journal of pragmatics, 35(9):1387–1411. [↗](https://doi.org/10.1016/s0378-2166(02)00182-0)
22. Roger J Kreuz and Gina M Caucci. 2007. [Lexical influences on the perception of sarcasm](https://md.research.elamur.ai/free/en/papers/kreuz-2007-lexical-influences-perception-sarcasm/). In Proceedings of the Workshop on computational approaches to Figurative Language, pages 1–4.
23. Shoushan Li, Sophia Yat Mei Lee, Ying Chen, Chu-Ren Huang, and Guodong Zhou. 2010. Sentiment classification and polarity shifting. In Proceedings of the 23rd International Conference on Computational Linguistics, pages 635–643. [find](https://scholar.google.com/scholar?q=Shoushan+Li%2C+Sophia+Yat+Mei+Lee%2C+Ying+Chen%2C+Chu%2DRen+Huang%2C+and+Guodong+Zhou%2E+2010%2E+Sentiment+classification+and+polarity+shifting%2E+In+Proceedings+of+the+23rd+International+Conference+on+Computational+)
24. Christine Liebrecht, Florian Kunneman, and Antal van den Bosch. 2013. [The perfect solution for detecting sarcasm in tweets# not](https://md.research.elamur.ai/free/en/papers/liebrecht-2013-perfect-solution-detecting-sarcasm-tweets/). In Proceedings of the 4th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis. New Brunswick, NJ: ACL.
25. George Miller and Christiane Fellbaum. 1998. Wordnet: An electronic lexical database. [find](https://scholar.google.com/scholar?q=George+Miller+and+Christiane+Fellbaum%2E+1998%2E+Wordnet%3A+An+electronic+lexical+database%2E)
26. Finn Årup Nielsen. 2011. A new anew: Evaluation of a word list for sentiment analysis in microblogs. In Proceedings of the ESWC2011 Workshop on ‘Making Sense of Microposts’: Big things come in small packages. [find](https://scholar.google.com/scholar?q=Finn+A%CC%8Arup+Nielsen%2E+2011%2E+A+new+anew%3A+Evaluation+of+a+word+list+for+sentiment+analysis+in+microblogs%2E+In+Proceedings+of+the+ESWC2011+Workshop+on+%E2%80%98Making+Sense+of+Microposts%E2%80%99%3A+Big+things+come+in+small+)
27. Bo Pang and Lillian Lee. 2008. Opinion mining and sentiment analysis. Foundations and trends in information retrieval, 2(1-2):1–135. [↗](https://doi.org/10.1561/1500000011)
28. Quintilian. 1922. With An English Translation. Harold Edgeworth Butler. Cambridge, Mass., Harvard University Press; London, William Heinemann, Ltd. [find](https://scholar.google.com/scholar?q=Quintilian%2E+1922%2E+With+An+English+Translation%2E+Harold+Edgeworth+Butler%2E+Cambridge%2C+Mass%2E%2C+Harvard+University+Press%3B+London%2C+William+Heinemann%2C+Ltd%2E)
29. Antonio Reyes, Paolo Rosso, and Tony Veale. 2013. A multidimensional approach for detecting irony in twitter. Language Resources and Evaluation, 47(1):239– 268. [↗](https://doi.org/10.1007/s10579-012-9196-x)
30. 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](https://md.research.elamur.ai/free/en/papers/riloff-2013-sarcasm-as-contrast-between-positive/). In Proceedings of Conference on Emperical Methods for Natural Language Processing (EMNLP), pages 704–714.
31. Angus Stevenson. 2010. Oxford dictionary of English. Oxford University Press. [find](https://scholar.google.com/scholar?q=Angus+Stevenson%2E+2010%2E+Oxford+dictionary+of+English%2E+Oxford+University+Press%2E)
32. Frank Stringfellow Jr. 1994. The meaning of irony: A psychoanalytic investigation. SUNY Press. [find](https://scholar.google.com/scholar?q=Frank+Stringfellow+Jr%2E+1994%2E+The+meaning+of+irony%3A+A+psychoanalytic+investigation%2E+SUNY+Press%2E)
33. Oren Tsur, Dmitry Davidov, and Ari Rappoport. 2010. Icwsm-a great catchy name: Semi-supervised recognition of sarcastic sentences in online product reviews. In In International AAAI Conference on Web and Social Media. [find](https://scholar.google.com/scholar?q=Oren+Tsur%2C+Dmitry+Davidov%2C+and+Ari+Rappoport%2E+2010%2E+Icwsm%2Da+great+catchy+name%3A+Semi%2Dsupervised+recognition+of+sarcastic+sentences+in+online+product+reviews%2E+In+In+International+AAAI+Conference+on+Web+)
34. Piyoros Tungthamthiti, Shirai Kiyoaki, and Masnizah Mohd. 2014. [Recognition of sarcasms in tweets based on concept level sentiment analysis and supervised learning approaches](https://md.research.elamur.ai/free/en/papers/tungthamthiti-2014-recognition-sarcasms-tweets-based-concept/). In Proceedings of Pacific Asia Conference on Language, Information and Computing, Phuket, Thailand.
35. Andrea Vanzo, Giuseppe Castellucci, Danilo Croce, and Roberto Basili. 2014. A context based model for sentiment analysis in twitter for the italian language. In R. Basili, A. Lenci, and B. Magnini, editors, Proceedings of the First Italian Conference on Computational Linguistics CLiC-it & the Fourth International Workshop EVALITA, pages 379–383, Pisa. Pisa University Press. [find](https://scholar.google.com/scholar?q=Andrea+Vanzo%2C+Giuseppe+Castellucci%2C+Danilo+Croce%2C+and+Roberto+Basili%2E+2014%2E+A+context+based+model+for+sentiment+analysis+in+twitter+for+the+italian+language%2E+In+R%2E+Basili%2C+A%2E+Lenci%2C+and+B%2E+Magnini%2C+ed)
36. Janyce Wiebe, Theresa Wilson, and Claire Cardie. 2005. Annotating expressions of opinions and emotions in language. Language resources and evaluation, 39(2-3):165–210. [↗](https://doi.org/10.1007/s10579-005-7880-9)
37. Gongjun Yan, Wu He, Jiancheng Shen, and Chuanyi Tang. 2014. A bilingual approach for conducting chinese and english social media sentiment analysis. Computer Networks, 75:491–503. [↗](https://doi.org/10.1016/j.comnet.2014.08.021)

*Updated 24 October 2026*