Advertisement

Using Feature Filtering Metrics as Meta-dimensions in Constructing Distributional Representations

  • Dongqiang YangEmail author
  • Yanqin Yin
  • Tonghui Han
  • Hongwei Ma
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11919)

Abstract

Feature filtering aims to find useful and relevant features for improvement of machine learning performance, reduction of computation complexity, and disclosure of internal information interaction. We employ some popular filtering criteria as meta-dimensions for the construction of feature space, where a word or a document can be represented with significantly reduced dimensionality. The experiment results show that the meta-feature data representation we proposed requires no extra resources on pre-training to derive word embeddings, and outperforms other traditional frequency-based or learning-based embeddings in the task of sentiment analysis.

Keywords

Feature selection Text classification Sentiment analysis 

Notes

Acknowledgement

This research was supported by the National Social Science Foundation of China (Grant No. 17BYY119) and the Humanity and Social Science Foundation of China Ministry of Education (Grant No. 15YJA740054).

References

  1. 1.
    Mladenić, D., Grobelnik, M.: Feature Selection for Unbalanced Class Distribution and Naive Bayes, pp. 258–267 (1999)Google Scholar
  2. 2.
    Forman, G.: An extensive empirical study of feature selection metrics for text classification. J. Mach. Learn. Res. 3, 1289–1305 (2003)zbMATHGoogle Scholar
  3. 3.
    Bengio, Y., et al.: A neural probabilistic language model. J. Mach. Learn. Res. 3, 1137–1155 (2003)zbMATHGoogle Scholar
  4. 4.
    Manning, C.D., Schütze, H.: Foundations of Statistical Natural Language Processing, 680 p. (1999)Google Scholar
  5. 5.
    Wiechmann, D.: On the computation of construction strength: testing measures of association as expressions of lexical bias. Corpus Linguist. Lingustic Theory 4, 253–290 (2008)Google Scholar
  6. 6.
    Chandrashekar, G., Sahin, F.: A survey on feature selection methods. Comput. Electr. Eng. 40(1), 16–28 (2014)CrossRefGoogle Scholar
  7. 7.
    Yang, Y., Pedersen, J.O.: A comparative study on feature selection in text categorization. In: ICML, pp. 412–420 (1997)Google Scholar
  8. 8.
    Mikolov, T., et al.: Distributed representations of words and phrases and their compositionality. In: NIPS, pp. 3111–3119 (2013)Google Scholar
  9. 9.
    Collobert, R., et al.: Natural language processing (almost) from scratch. J. Mach. Learn. Res. 12, 2493–2537 (2011)zbMATHGoogle Scholar
  10. 10.
    Pennington, J., Socher, R., Manning, C.: Glove: Global Vectors for Word Representation, vol. 14, pp. 1532–1543 (2014)Google Scholar
  11. 11.
    Levy, O., Goldberg, Y.: Neural word embedding as implicit matrix factorization. In: NIPS, pp. 2177–2185 (2014)Google Scholar
  12. 12.
    Baroni, M., Dinu, G., Kruszewski, G.: Don’t count, predict! A systematic comparison of context-counting vs. context-predicting semantic vectors. In: ACL, pp. 238–247 (2014)Google Scholar
  13. 13.
    Faruqui, M., et al.: Retrofitting Word Vectors to Semantic Lexicons (2014)Google Scholar
  14. 14.
    Yu, M., Dredze, M.: Improving lexical embeddings with semantic knowledge. In: ACL, pp. 545–550 (2014)Google Scholar
  15. 15.
    Tang, D., et al.: Learning sentiment-specific word embedding for Twitter sentiment classification. In: ACL, pp. 1555–1565 (2014)Google Scholar
  16. 16.
    Kim, Y.: Convolutional Neural Networks for Sentence Classification (2014)Google Scholar
  17. 17.
    Conneau, A., et al.: Very deep convolutional networks for text classification. In: EACL, pp. 1107–1116 (2017)Google Scholar
  18. 18.
    Barnes, J., Klinger, R., Walde, S.S.I.: Assessing State-of-the-Art Sentiment Models on State-of-the-Art Sentiment Datasets (2017)Google Scholar
  19. 19.
    Zhang, L., Wang, S., Liu, B.: Deep learning for sentiment analysis: a survey. Wiley Interdisc. Rev.: Data Mining Knowl. Discov. 8(4), e1253 (2018)Google Scholar
  20. 20.
    McCann, B., et al.: Learned in translation: contextualized word vectors. In: NIPS, pp. 6297–6308 (2017)Google Scholar
  21. 21.
    Peters, M.E., et al.: Deep contextualized word representations. In: NAACL 2018, pp. 2227–2237 (2018)Google Scholar
  22. 22.
    Howard, J., Ruder, S.: Universal language model fine-tuning for text classification. In: ACL (2018)Google Scholar
  23. 23.
    Devlin, J., et al.: BERT: pre-training of deep bidirectional transformers for language understanding. In: NAACL-HLT (2018)Google Scholar
  24. 24.
    Zheng, Z., Xiaoyun, W., Srihari, R.: Feature selection for text categorization on imbalanced data. SIGKDD Explor. Newsl. 6(1), 80–89 (2004)CrossRefGoogle Scholar
  25. 25.
    Zipf, G.K.: Human behavior and the principle of least effort: an introduction to human ecology. Facsim. of 1949 ed, 573 p. (1965)Google Scholar
  26. 26.
    Jones, K.S.: Automatic summarising: factors and directions. In: Advances in Automatic Text Summarization (1998)Google Scholar
  27. 27.
    Deerwester, S.C., et al.: Indexing by latent semantic analysis. J. Am. Soc. Inf. Sci. 41(6), 391–407 (1990)CrossRefGoogle Scholar
  28. 28.
    Socher, R., et al.: Parsing natural scenes and natural language with recursive neural networks. In: ICML, pp. 129–136 (2011)Google Scholar
  29. 29.
    Maas, A.L., et al.: Learning word vectors for sentiment analysis. In: Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, vol. 1, pp. 142–150 (2011)Google Scholar
  30. 30.
    Tang, D., Qin, B., Liu, T.: Document modeling with gated recurrent neural network for sentiment classification. In: EMNLP, pp. 1422–1432 (2015)Google Scholar
  31. 31.
    Wilson, T., Wiebe, J., Hoffmann, P.: Recognizing contextual polarity in phrase-level sentiment analysis. In: HLT-EMNLP, pp. 347–354 (2005)Google Scholar
  32. 32.
    Mikolov, T., et al.: Efficient Estimation of Word Representations in Vector Space (2013)Google Scholar
  33. 33.
    Radford, A., et al.: Improving language understanding by generative pre-training (2018)Google Scholar
  34. 34.
    Le, Q., Mikolov, T.: Distributed representations of sentences and documents. In: ICML, pp. 1188–1196 (2014)Google Scholar
  35. 35.
    Johnson, R., Zhang, T.: Supervised and semi-supervised text categorization using LSTM for region embeddings. In: ICML, pp. 526–534 (2016)Google Scholar
  36. 36.
    Miyato, T., Dai, A.M., Goodfellow, I.: Adversarial Training Methods for Semi-Supervised Text Classification (2017)Google Scholar
  37. 37.
    Xie, Q., et al.: Unsupervised Data Augmentation for Consistency Training. eprint arXiv:190412848 (2019)
  38. 38.
    Liu, Y., Lapata, M.: Learning structured text representations. Trans. Assoc. Comput. Linguist. 6, 63–75 (2018)CrossRefGoogle Scholar
  39. 39.
    Yang, Z., et al.: Hierarchical attention networks for document classification. In: HLT-NAACL, pp. 1480–1489 (2016)Google Scholar
  40. 40.
    Melis, G., Dyer, C., Blunsom, P.: On the state of the art of evaluation in neural language models. In: International Conference on Learning Representations (2018)Google Scholar
  41. 41.
    Zilly, J.G., et al.: Recurrent highway networks. In: ICML, pp. 4189–4198 (2017)Google Scholar
  42. 42.
    Zoph, B., Le, Q.V.: Neural Architecture Search with Reinforcement Learning. arXiv e-prints arXiv:161101578Z (2016)
  43. 43.
    Lipton, Z.C., Steinhardt, J.: Troubling Trends in Machine Learning Scholarship. arXiv e-prints arXiv:180703341L (2018)
  44. 44.
    Frankle, J., Carbin, M.: The lottery ticket hypothesis: finding sparse, trainable neural networks. In: ICLR (2019)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Dongqiang Yang
    • 1
    Email author
  • Yanqin Yin
    • 1
  • Tonghui Han
    • 2
  • Hongwei Ma
    • 1
  1. 1.School of Computer Science and TechnologyShandong Jianzhu UniversityJinanChina
  2. 2.Binzhou Polytechnic CollegeBinzhouChina

Personalised recommendations