Cognitive Computation

, Volume 7, Issue 2, pp 186–197 | Cite as

Propagating and Aggregating Fuzzy Polarities for Concept-Level Sentiment Analysis

  • Mauro Dragoni
  • Andrea G. B. Tettamanzi
  • Célia da Costa Pereira


An emerging field within sentiment analysis concerns the investigation about how sentiment polarities associated with concepts have to be adapted with respect to the different domains in which they are used. In this paper, we explore the use of fuzzy logic for modeling concept polarities, and the uncertainty associated with them, with respect to different domains. The approach is based on the use of a knowledge graph built by combining two linguistic resources, namely WordNet and SenticNet. Such a knowledge graph is then exploited by a graph-propagation algorithm that propagates sentiment information learned from labeled datasets. The system implementing the proposed approach has been evaluated on the Blitzer dataset. The results demonstrate its viability in real-world cases.


Sentiment analysis Multi-domain learning Fuzzy logic 


  1. 1.
    Pang B, Lee L, Vaithyanathan S. Thumbs up? sentiment classification using machine learning techniques. In: Proceedings of the conference on empirical methods in natural language processing (EMNLP), Philadelphia, Association for Computational Linguistics; July 2002. p. 79–86.Google Scholar
  2. 2.
    Liu B, Zhang L. A survey of opinion mining and sentiment analysis. In: Aggarwal CC, Zhai CX, editors. Mining text data. Berlin: Springer; 2012. p. 415–63.CrossRefGoogle Scholar
  3. 3.
    Blitzer J, Dredze M, Pereira F. Biographies, bollywood, boom-boxes and blenders: domain adaptation for sentiment classification. In: ACL; 2007. p. 187–205.Google Scholar
  4. 4.
    Pang B, Lee L. Opinion mining and sentiment analysis. Found Trends Inf Retr. 2008;2(1–2):1–135.CrossRefGoogle Scholar
  5. 5.
    Pang B, Lee L. A sentimental education: sentiment analysis using subjectivity summarization based on minimum cuts. In: ACL; 2004. p. 271–78.Google Scholar
  6. 6.
    Dave K, Lawrence S, Pennock DM. Mining the peanut gallery: opinion extraction and semantic classification of product reviews. In: WWW; 2003. p. 519–28.Google Scholar
  7. 7.
    Paltoglou G, Thelwall M. A study of information retrieval weighting schemes for sentiment analysis. In: ACL; 2010. p. 1386–95.Google Scholar
  8. 8.
    Tan S, Wang Y, Cheng X. Combining learn-based and lexicon-based techniques for sentiment detection without using labeled examples. In: SIGIR; 2008. p. 743–44.Google Scholar
  9. 9.
    Qiu L, Zhang W, Hu C, Zhao K. Selc: a self-supervised model for sentiment classification. In: CIKM; 2009. p. 929–36.Google Scholar
  10. 10.
    Melville P, Gryc W, Lawrence RD. Sentiment analysis of blogs by combining lexical knowledge with text classification. In: KDD; 2009. p. 1275–284.Google Scholar
  11. 11.
    Taboada M, Brooke J, Tofiloski M, Voll KD, Stede M. Lexicon-based methods for sentiment analysis. Comput Linguist. 2011;37(2):267–307.CrossRefGoogle Scholar
  12. 12.
    Turney PD. Thumbs up or thumbs down? Semantic orientation applied to unsupervised classification of reviews. In: ACL; 2002. p. 417–24.Google Scholar
  13. 13.
    Somasundaran S. Discourse-level relations for opinion analysis. PhD thesis, University of Pittsburgh; 2010.Google Scholar
  14. 14.
    Asher N, Benamara F, Mathieu YY. Distilling opinion in discourse: a preliminary study. In: COLING (Posters); 2008. p. 7–10.Google Scholar
  15. 15.
    Wang H, Zhou G. Topic-driven multi-document summarization. In: IALP; 2010. p. 195–98.Google Scholar
  16. 16.
    Riloff E, Patwardhan S, Wiebe J. Feature subsumption for opinion analysis. In: EMNLP; 2006. p. 440–48.Google Scholar
  17. 17.
    Wiebe J, Wilson T, Bruce RF, Bell M, Martin M. Learning subjective language. Comput Linguist. 2004;30(3):277–308.CrossRefGoogle Scholar
  18. 18.
    Wilson T, Wiebe J, Hwa R. Just how mad are you? Finding strong and weak opinion clauses. In: AAAI; 2004. p. 761–69.Google Scholar
  19. 19.
    Wilson T, Wiebe J, Hwa R. Recognizing strong and weak opinion clauses. Comput Intell. 2006;22(2):73–99.CrossRefGoogle Scholar
  20. 20.
    Yu H, Hatzivassiloglou V. Towards answering opinion questions: separating facts from opinions and identifying the polarity of opinion sentences. In: Proceedings of the 2003 conference on empirical methods in natural language processing. EMNLP ’03. Stroudsburg, PA: Association for Computational Linguistics; 2003. p. 129–36.Google Scholar
  21. 21.
    Hatzivassiloglou V, Wiebe J. Effects of adjective orientation and gradability on sentence subjectivity. In: COLING; 2000. p. 299–305.Google Scholar
  22. 22.
    Kim SM, Hovy EH. Crystal: Analyzing predictive opinions on the web. In: EMNLP-CoNLL; 2007. p. 1056–64.Google Scholar
  23. 23.
    Kim SM, Pantel P, Chklovski T, Pennacchiotti M. Automatically assessing review helpfulness. In: EMNLP; 2006. p. 423–30.Google Scholar
  24. 24.
    Jakob N, Gurevych I. Extracting opinion targets in a single and cross-domain setting with conditional random fields. In: EMNLP; 2010. p. 1035–45.Google Scholar
  25. 25.
    Lafferty JD, McCallum A, Pereira FCN. Conditional random fields: probabilistic models for segmenting and labeling sequence data. In: ICML; 2001. p. 282–89.Google Scholar
  26. 26.
    Freitag D, McCallum A. Information extraction with HMM structures learned by stochastic optimization. In: AAAI/IAAI; 2000. p. 584–89.Google Scholar
  27. 27.
    Jin W, Ho HH. A novel lexicalized HMM-based learning framework for web opinion mining. In: Proceedings of the 26th annual international conference on machine learning, ICML ’09. New York, NY: ACM; 2009. p. 465–72.Google Scholar
  28. 28.
    Jin W, Ho HH, Srihari RK. Opinionminer: a novel machine learning system for web opinion mining and extraction. In: KDD; 2009. p. 1195–1204.Google Scholar
  29. 29.
    Liu B, Hu M, Cheng J. Opinion observer: analyzing and comparing opinions on the web. In: WWW; 2005. p. 342–51.Google Scholar
  30. 30.
    Wu Y, Zhang Q, Huang X, Wu L. Phrase dependency parsing for opinion mining. In: EMNLP; 2009. p. 1533–41.Google Scholar
  31. 31.
    Su Q, Xu X, Guo H, Guo Z, Wu X, Zhang X, Swen B, Su Z. Hidden sentiment association in chinese web opinion mining. In: WWW; 2008. p. 959–68.Google Scholar
  32. 32.
    Qiu G, Liu B, Bu J, Chen C. Expanding domain sentiment lexicon through double propagation. In: IJCAI; 2009. p. 1199–1204.Google Scholar
  33. 33.
    Qiu G, Liu B, Bu J, Chen C. Opinion word expansion and target extraction through double propagation. Comput Linguist. 2011;37(1):9–27.CrossRefGoogle Scholar
  34. 34.
    Barbosa L, Feng J. Robust sentiment detection on twitter from biased and noisy data. In: COLING (Posters); 2010. p. 36–44.Google Scholar
  35. 35.
    Bermingham A, Smeaton AF. Classifying sentiment in microblogs: is brevity an advantage? In: CIKM; 2010. p. 1833–36.Google Scholar
  36. 36.
    Go A, Bhayani R, Huang L. Twitter sentiment classification using distant supervision. CS224N Project Report, Standford University; 2009.Google Scholar
  37. 37.
    Cambria E, Hussain A. Sentic computing: techniques, tools, and applications. Volume 2 of Springerbriefs in cognitive computation. Dordrecht: Springer; 2012.CrossRefGoogle Scholar
  38. 38.
    Cambria E, Hussain A. Sentic album: content-, concept-, and context-based online personal photo management system. Cognit Comput. 2012;4(4):477–96.CrossRefGoogle Scholar
  39. 39.
    Wang QF, Cambria E, Liu CL, Hussain A. Common sense knowledge for handwritten chinese recognition. Cognit Comput. 2013;5(2):234–42.CrossRefGoogle Scholar
  40. 40.
    Aue A, Gamon M. Customizing sentiment classifiers to new domains: a case study. In: Proceedings of RANLP; 2005.Google Scholar
  41. 41.
    Yang H, Callan J, Si L. Knowledge transfer and opinion detection in the TREC 2006 blog track. In: TREC; 2006.Google Scholar
  42. 42.
    Pan SJ, Ni X, Sun JT, Yang Q, Chen Z. Cross-domain sentiment classification via spectral feature alignment. In: WWW; 2010. p. 751–60.Google Scholar
  43. 43.
    Bollegala D, Weir DJ, Carroll JA. Cross-domain sentiment classification using a sentiment sensitive thesaurus. IEEE Trans Knowl Data Eng. 2013;25(8):1719–31.CrossRefGoogle Scholar
  44. 44.
    Xia R, Zong C, Hu X, Cambria E. Feature ensemble plus sample selection: domain adaptation for sentiment classification. IEEE Int Syst. 2013;28(3):10–8.CrossRefGoogle Scholar
  45. 45.
    Yoshida Y, Hirao T, Iwata T, Nagata M, Matsumoto Y. Transfer learning for multiple-domain sentiment analysis—identifying domain dependent/independent word polarity. In: AAAI; 2011. p. 1286–91.Google Scholar
  46. 46.
    Ponomareva N, Thelwall M. Semi-supervised vs. cross-domain graphs for sentiment analysis. In: RANLP; 2013. p. 571–78.Google Scholar
  47. 47.
    Tsai ACR, Wu CE, Tsai RTH, Jen Hsu JY. Building a concept-level sentiment dictionary based on commonsense knowledge. IEEE Int Syst. 2013;28(2):22–30.CrossRefGoogle Scholar
  48. 48.
    Tai YJ, Kao HY. Automatic domain-specific sentiment lexicon generation with label propagation. In: iiWAS, ACM; 2013. p. 53:53–53:62.Google Scholar
  49. 49.
    Huang S, Niu Z, Shi C. Automatic construction of domain-specific sentiment lexicon based on constrained label propagation. Knowl Based Syst. 2014;56:191–200.CrossRefGoogle Scholar
  50. 50.
    Zadeh LA. Fuzzy sets. Inf Control. 1965;8:338–53.CrossRefGoogle Scholar
  51. 51.
    Zadeh LA. Fuzzy sets as a basis for a theory of possibility. Fuzzy Sets Syst. 1978;1:3–28.CrossRefGoogle Scholar
  52. 52.
    Zadeh LA. The concept of a linguistic variable and its application to approximate reasoning—I. Inf Sci. 1975;8(3):199–249.CrossRefGoogle Scholar
  53. 53.
    Hellendoorn H, Thomas C. Defuzzification in fuzzy controllers. Intell Fuzzy Syst. 1993;1:109–23.Google Scholar
  54. 54.
    Fellbaum C. WordNet: an electronic lexical database. Cambridge: MIT Press; 1998.Google Scholar
  55. 55.
    Cambria E, Olsher D, Rajagopal D. Senticnet 3: a common and common-sense knowledge base for cognition–driven sentiment analysis. In: AAAI; 2014. p. 1515–21.Google Scholar
  56. 56.
    Baccianella S, Esuli A, Sebastiani F. Sentiwordnet 3.0: an enhanced lexical resource for sentiment analysis and opinion mining. In: LREC; 2010. p. 2200–04.Google Scholar
  57. 57.
    Grassi M. Developing heo human emotions ontology. In: COST 2101/2102 conference; 2009. p. 244–51.Google Scholar
  58. 58.
    Dong W, Shah H, Wong F. Fuzzy computations in risk and decision analysis. Civ Eng Syst. 1985;2:201–8.CrossRefGoogle Scholar
  59. 59.
    Kirkpatrick S, Gelatt CD, Vecchi MP. Optimization by simulated annealing. Science. 1983;220(4598):671–80.CrossRefPubMedGoogle Scholar
  60. 60.
    Chang CC, Lin CJ. Libsvm: a library for support vector machines. ACM TIST. 2011;2(3):27:1–27.Google Scholar
  61. 61.
    McCallum AK. Mallet: a machine learning for language toolkit. (2002).

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Mauro Dragoni
    • 1
  • Andrea G. B. Tettamanzi
    • 2
  • Célia da Costa Pereira
    • 2
  1. 1.FBK–IRSTTrentoItaly
  2. 2.I3S, UMR 7271Université Nice Sophia AntipolisNiceFrance

Personalised recommendations