Advertisement

Semi-supervised Sentiment Classification Based on Auxiliary Task Learning

  • Huan Liu
  • Jingjing Wang
  • Shoushan LiEmail author
  • Junhui Li
  • Guodong Zhou
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11109)

Abstract

Sentiment classification is an important task in the community of Nature Language Processing. This task aims to determine the sentiment category towards a piece of text. One challenging problem of this task is that it is difficult to obtain a large number of labeled samples. Therefore, a large number of studies are focused on semi-supervised learning, i.e., learning information from unlabeled samples. However, one disadvantage of the previous methods is that the unlabeled samples and the labeled samples are studied in different models, and there is no interaction between them. Based on this, this paper tackles the problem by proposing a semi-supervised sentiment classification based on auxiliary task learning, namely Aux-LSTM, which is used to assist learning the sentiment classification task with a small amount of human-annotated samples by training auto-annotated samples. Specifically, the two tasks are allowed to share the auxiliary LSTM layer, and the auxiliary expression obtained by the auxiliary LSTM layer is used to assist the main task. Empirical studies demonstrate that the proposed method can effectively improve the experimental performance.

Keywords

Sentiment classification Auxiliary task Auto-annotation samples 

Notes

Acknowledgments

This research work has been partially supported by two NSFC grants, No. 61672366 and No. 61331011.

References

  1. 1.
    Ye, Q., Zhang, Z., Law, R.: Sentiment classification of online reviews to travel destinations by supervised machine learning approaches. Expert Syst. Appl. 36(3), 6527–6535 (2009)CrossRefGoogle Scholar
  2. 2.
    Pang, B., Lee, L., Vaithyanathan, S.: Thumbs up? Sentiment classification using machine learning techniques. In: Proceedings of EMNLP, Philadelphia, pp. 79–86 (2002)Google Scholar
  3. 3.
    Wan, X.: Co-training for cross-lingual sentiment classification. In: Proceedings of ACL-IJCNLP, Singapore, pp. 235–243 (2009)Google Scholar
  4. 4.
    Zhu, X., Ghahramani, Z.: Learning from Labeled and Unlabeled Data with Label Propagation. CMU CALD Technical report, CMU-CALD-02-107Google Scholar
  5. 5.
    Zhou, S., Chen, Q., Wang, X.: Active deep networks for semi-supervised sentiment classification. In: Proceedings of COLING, Beijing, pp. 1515–1523 (2010)Google Scholar
  6. 6.
    Li, S., Xue, Y., Wang, Z., et al.: Active learning for cross-domain sentiment classification. In: Proceedings of IJCAI, Beijing, pp. 2127–2133 (2013)Google Scholar
  7. 7.
    Xia, R., Wang, C., Dai, X., et al.: Co-training for semi-supervised sentiment classification based on dual-view bags-of-words representation. In: Proceedings of ACL-IJCNLP, Beijing, pp. 1054–1063 (2015)Google Scholar
  8. 8.
    Sharma, R., Somani, A., Kumar, L., et al.: Sentiment intensity ranking among adjectives using sentiment bearing word embeddings. In: Proceddings of EMNLP, pp. 547–552 (2017)Google Scholar
  9. 9.
    Yu, J., Jiang, J.: Learning sentence embeddings with auxiliary tasks for cross-domain sentiment classification. In: Proceddings of EMNLP, pp. 236–246 (2016)Google Scholar
  10. 10.
    Shi, H., Jia, D., Miao, P.: Improved information gain text feature selection algorithm based on word frequency information. J. Comput. Appl. 34, 3279–3282 (2014)Google Scholar
  11. 11.
    Xu, J., Jiang, H.: An improved information gain feature selection algorithm for SVM text classifier. In: Proceddings of IEEE, pp. 273–276 (2015)Google Scholar
  12. 12.
    Zhang, H., Ren, Y.G., Yang, X.: Research on text feature selection algorithm based on information gain and feature relation tree. In: Proceddings of IEEE, pp. 446–449 (2014)Google Scholar
  13. 13.
    Hochreiter, K., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  14. 14.
    Graves, A.: Generating sequences with recurrent neural networks. Computer Science (2013)Google Scholar
  15. 15.
    Glorot, X., Bordes, A., Bengio, Y.: Deep sparse rectifier neural networks. In: JMLR W&CP, vol. 15, pp. 315–323 (2012)Google Scholar
  16. 16.
    Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. Computer Science (2014)Google Scholar
  17. 17.
    Blitzer, J., Dredze, M., Pereira, F.: Biographies, bollywood, boom-boxes and blenders: domain adaptation for sentiment classification. In: Proceedings of ACL, Prague, pp. 440–447 (2007)Google Scholar
  18. 18.
    Li, S., Huang, L., Wang, J., et al.: Semi-stacking for semi-supervised sentiment classification. In: Proceedings of ACL-IJCNLP, Beijing, pp. 27–31 (2015)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Huan Liu
    • 1
  • Jingjing Wang
    • 1
  • Shoushan Li
    • 1
    Email author
  • Junhui Li
    • 1
  • Guodong Zhou
    • 1
  1. 1.Natural Language Processing Lab, School of Computer Science and TechnologySoochow UniversitySuzhouChina

Personalised recommendations