Joint Naïve Bayes and LDA for Unsupervised Sentiment Analysis

  • Yong Zhang
  • Dong-Hong Ji
  • Ying Su
  • Hongmiao Wu
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7818)


In this paper we proposed a hierarchical generative model based on Naïve Bayes and LDA for unsupervised sentiment analysis at sentence level and document level of granularity simultaneously. In particular, our model called NB-LDA assumes that each sentence instead of word has a latent sentiment label, and then the sentiment label generates a series of features for the sentence independently in the Naïve Bayes manner. The idea of NB assumption at sentence level makes it possible that we can use advanced NLP technologies such as dependency parsing to improve the performance for unsupervised sentiment analysis. Experiment results show that the proposed NB-LDA can obtain significantly improved results for sentiment analysis comparing to other approaches.


Sentiment Analysis Latent Dirichlet Allocation Naïve Bayes Opinion Mining 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Banerjee, A., Shan, H.: Latent Dirichlet conditional Naive-Bayes models. In: The IEEE International Conference on Data Mining (ICDM), pp. 421–426 (2007)Google Scholar
  2. 2.
    Blei, D., McAuliffe, J.: Supervised topic models. In: Proceeding of Neural Information Processing Systems, NIPS (2007)Google Scholar
  3. 3.
    Blei, D., Ng, A., Jordan, M.: Latent Dirichlet allocation. Journal of Machine Learning Research 3(5), 993–1022 (2003)zbMATHGoogle Scholar
  4. 4.
    Dasgupta, S., Ng, V.: Topic-wise, Sentiment-wise, or Otherwise? Identifying the Hidden Dimension for Unsupervised Text Classification. In: Proceedings of EMNLP (2009)Google Scholar
  5. 5.
    Ding, X., Liu, B., Zhang, L.: Entity discovery and assignment for opinion mining applications. In: Proceedings of KDD 2009, pp. 1125–1134 (2009)Google Scholar
  6. 6.
    Griffiths, T., Steyvers, M.: Finding scientific topics. Proceedings of the National Academy of Sciences 101(90001), 5228–5235 (2004)CrossRefGoogle Scholar
  7. 7.
    Gruber, A., Rosen-Zvi, M., Weiss, Y.: Hidden Topic Markov Models. In: Artificial Intelligence and Statistics (AISTATS), San Juan, Puerto Rico (2007)Google Scholar
  8. 8.
    Jo, Y., Oh, A.H.: Aspect and sentiment unification model for online review analysis. In: Proceedings of WSDM 2011, pp. 815–824 (2011)Google Scholar
  9. 9.
    Li, T., Zhang, Y., Sindhwani, V.: A nonnegative matrix tri-factorization approach to sentiment classification with lexical prior knowledge. In: Proceedings of ACL-IJCNLP 2009, pp. 244–252 (2009)Google Scholar
  10. 10.
    Lin, C., He, Y.: Joint sentiment/topic model for sentiment analysis. In: Proceeding of the Conference on Information and Knowledge Management, CIKM (2009)Google Scholar
  11. 11.
    McDonald, R., Hannan, K., Neylon, T., Wells, M., Reynar, J.: Structured models for fine-to-coarse sentiment analysis. In: Proceedings of ACL (2007)Google Scholar
  12. 12.
    Mei, Q., Ling, X., Wondra, M., Su, H., Zhai, C.X.: Topic sentiment mixture: modeling facets and opinions in weblogs. In: Proceedings of WWW (2007)Google Scholar
  13. 13.
    Mimno, D.M., McCallum, A.: Topic Models Conditioned on Arbitrary Features with Dirichlet-multinomial Regression. In: Proceedings of the 24th Conference in Uncertainty in Artificial Intelligence, pp. 411–418 (2008)Google Scholar
  14. 14.
    Mukherjee, A.: Liu, Bing: Modeling Review Comments. In: Proceedings of ACL 2012, Jeju, Republic of Korea, July 8-14 (2012)Google Scholar
  15. 15.
    Pang, B., Lee, L.: A sentimental education: Sentiment analysis using subjectivity summarization based on minimum cuts. In: Proceedings of ACL (2004)Google Scholar
  16. 16.
    Täckström, O., McDonald, R.: Discovering fine-grained sentiment with latent variable structured prediction models. In: Clough, P., Foley, C., Gurrin, C., Jones, G.J.F., Kraaij, W., Lee, H., Mudoch, V. (eds.) ECIR 2011. LNCS, vol. 6611, pp. 368–374. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  17. 17.
    Täckström, O., McDonald, R.: Semi-supervised Latent Variable Models for Sentence-level Sentiment Analysis. In: Proceedings of ACL (2011)Google Scholar
  18. 18.
    Wilson, T., Wiebe, J., Hoffmann, P.: Recognizing contextual polarity in phrase-level sentiment analysis. In: Proceedings of EMNLP (2005)Google Scholar
  19. 19.
    Zhang, Y., Ji, D.-H., Su, Y., Sun, C.: Sentiment Analysis for Online Reviews Using an Author-Review-Object Model. In: Salem, M.V.M., Shaalan, K., Oroumchian, F., Shakery, A., Khelalfa, H. (eds.) AIRS 2011. LNCS, vol. 7097, pp. 362–371. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  20. 20.
    Zhao, W.X., Jiang, J., Yan, H., Li, X.: Jointly modeling aspects and opinions with a MaxEnt-LDA hybrid. In: Proceedings of EMNLP 2010, Stroudsburg, PA, USA, pp. 56–65 (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Yong Zhang
    • 1
    • 2
  • Dong-Hong Ji
    • 1
  • Ying Su
    • 3
  • Hongmiao Wu
    • 4
  1. 1.Computer SchoolWuhan UniversityWuhanP.R. China
  2. 2.Department of Computer ScienceHuazhong Normal UniversityWuhanP.R. China
  3. 3.Department of Computer Science, Wuchang BranchHuazhong University of Science and TechnologyWuhanP.R. China
  4. 4.School of Foreign Languages and LiteratureWuhan UniversityWuhanP.R. China

Personalised recommendations