Regularising LSTM classifier by transfer learning for detecting misogynistic tweets with small training set


Supervised machine learning methods depend highly on the quality of the training dataset and the underlying model. In particular, neural network models, that have shown great success in dealing with natural language problems, require a large dataset to learn a vast number of parameters. However, it is not always easy to build a large (labelled) dataset. For example, due to the complex nature of tweets and the manual labour involved, it is hard to create a large Twitter data set with the misogynistic label. In this paper, we propose to regularise a long short-term memory (LSTM) classifier using a pretrained LSTM-based language model (LM) to build an accurate classification model with a small training set. We explain transfer learning (TL) with a Bayesian interpretation and show that TL can be viewed as an uncertainty regularisation technique in Bayesian inference. We show that a LM pre-trained on a sequence of general to task-specific domain datasets can be used to regularise a LSTM classifier effectively when a small training dataset is available. Empirical analysis with two small Twitter datasets reveals that an LSTM model trained in this way can outperform the state-of-the-art classification models.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3


  1. 1.

  2. 2.


  1. 1.

    Ahluwalia R, Soni H, Callow E, Nascimento A, De Cock M (2018) Detecting hate speech against women in English tweets. EVALITA Eval NLP Speech Tools Ital 12:194

    Article  Google Scholar 

  2. 2.

    Amnesty International (2018) Toxic twitter—a toxic place for women.

  3. 3.

    Badjatiya P, Gupta S, Gupta M, Varma V (2017) Deep learning for hate speech detection in tweets. In: Proceedings of the 26th international conference on world wide web companion, international world wide web conferences steering committee, pp 759–760

  4. 4.

    Bartlett J, Norrie R, Patel S, Rumpel R, Wibberley S (2014) Misogyny on twitter. Demos. Retrieved from analysis and policy observatory website

  5. 5.

    Bashar MA, Nayak R, Suzor N, Weir B (2018) Misogynistic tweet detection: modelling CNN with small datasets. In: The 16th Australasian data mining conference

  6. 6.

    Blundell C, Cornebise J, Kavukcuoglu K, Wierstra D (2015) Weight uncertainty in neural network. In: International conference on machine learning, pp 1613–1622

  7. 7.

    Bouchard G, Triggs B (2004) The tradeoff between generative and discriminative classifiers. In: 16th IASC international symposium on computational statistics (COMPSTAT’04), pp 721–728

  8. 8.

    Bradbury J, Merity S, Xiong C, Socher R (2016) Quasi-recurrent neural networks. arXiv preprint arXiv:1611.01576

  9. 9.

    Chen T, Guestrin C (2016) Xgboost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. ACM, pp 785–794

  10. 10.

    Dai AM, Le QV (2015) Semi-supervised sequence learning. In: Advances in neural information processing systems, pp 3079–3087

  11. 11.

    Davidson T, Warmsley D, Macy M, Weber I (2017) Automated hate speech detection and the problem of offensive language. arXiv preprint arXiv:1703.04009

  12. 12.

    Devlin J, Chang M-W, Lee K, Toutanova K (2018) Bert: pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805

  13. 13.

    Downey A (2012) Think Bayes: Bayesian statistics made simple. Green Tea Press, Needham

    Google Scholar 

  14. 14.

    Dragiewicz M, Burgess J, Matamoros-Fernández A, Salter M, Suzor NP, Woodlock D, Harris B (2018) Technology facilitated coercive control: domestic violence and the competing roles of digital media platforms. Fem Med Stud 18:1–17

    Article  Google Scholar 

  15. 15.

    Fadaee M, Bisazza A, Monz C (2017) Data augmentation for low-resource neural machine translation. In: Proceedings of the 55th annual meeting of the association for computational linguistics (volume 2: short papers), vol 2, pp 567–573

  16. 16.

    Fersini E, Nozza D, Rosso P (2018) Overview of the evalita 2018 task on automatic misogyny identification (AMI). In: Proceedings of the 6th evaluation campaign of Natural Language Processing and Speech tools for Italian (EVALITA’18). Italy, Turin

  17. 17.

    Gal Y (2016) Uncertainty in deep learning. University of Cambridge, Cambridge

    Google Scholar 

  18. 18.

    Gal Y, Ghahramani Z (2016) A theoretically grounded application of dropout in recurrent neural networks. In: Advances in neural information processing systems, pp 1019–1027

  19. 19.

    Gitari ND, Zuping Z, Damien H, Long J (2015) A lexicon-based approach for hate speech detection. Int J Multimed Ubiquitous Eng 10(4):215–230

    Article  Google Scholar 

  20. 20.

    Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp 249–256

  21. 21.

    Hearst MA, Dumais ST, Osuna E, Platt J, Scholkopf B (1998) Support vector machines. IEEE Intell Syst Appl 13(4):18–28

    Article  Google Scholar 

  22. 22.

    Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780

    Article  Google Scholar 

  23. 23.

    Hoerl AE, Kennard RW (1970) Ridge regression: applications to nonorthogonal problems. Technometrics 12(1):69–82

    Article  Google Scholar 

  24. 24.

    Howard J, Ruder S (2018) Universal language model fine-tuning for text classification. In: Proceedings of the 56th annual meeting of the association for computational linguistics (volume 1: long papers), vol 1, pp 328–339

  25. 25.

    Jozefowicz R, Vinyals O, Schuster M, Shazeer N, Wu Y (2016) Exploring the limits of language modeling. arXiv preprint arXiv:1602.02410

  26. 26.

    Kim Y (2014) Convolutional neural networks for sentence classification. In: Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), pp 1746–1751

  27. 27.

    Kwok I, Wang Y (2013) Locate the hate: detecting tweets against blacks. In: Twenty-seventh AAAI conference on artificial intelligence

  28. 28.

    Lee DD, Seung HS (2001) Algorithms for non-negative matrix factorization. In: Advances in neural information processing systems, pp 556–562

  29. 29.

    Lewis DD (1998) Naive (Bayes) at forty: the independence assumption in information retrieval. In: European conference on machine learning. Springer, pp 4–15

  30. 30.

    Li Y, Algarni A, Zhong N (2010) Mining positive and negative patterns for relevance feature discovery. In: Proceedings of the 16th ACM SIGKDD international conference on knowledge discovery and data mining. ACM, Washington, pp 753–762

  31. 31.

    Li Y, Gal Y (2017) Dropout inference in bayesian neural networks with alpha-divergences. In: Proceedings of the 34th international conference on machine learning,, vol 70, pp 2052–2061

  32. 32.

    Liaw A, Wiener M et al (2002) Classification and regression by randomforest. R News 2(3):18–22

    Google Scholar 

  33. 33.

    Liu P, Li W, Zou L (2019) Nuli at semeval-2019 task 6: transfer learning for offensive language detection using bidirectional transformers. In: Proceedings of the 13th international workshop on semantic evaluation, pp 87–91

  34. 34.

    Logeswaran L, Lee H (2018) An efficient framework for learning sentence representations. arXiv preprint arXiv:1803.02893

  35. 35.

    Maas AL, Daly RE, Pham PT, Huang D, Ng AY, Potts C (2011) Learning word vectors for sentiment analysis. In: Proceedings of the 49th annual meeting of the association for computational linguistics: human language technologies. Association for Computational Linguistics, vol 1, pp 142–150

  36. 36.

    MacKay DJ (1992) A practical bayesian framework for backpropagation networks. Neural Comput 4(3):448–472

    Article  Google Scholar 

  37. 37.

    McHugh ML (2012) Interrater reliability: the kappa statistic. Biochem Med 22(3):276–282

    MathSciNet  Article  Google Scholar 

  38. 38.

    Melis G, Dyer C, Blunsom P (2017) On the state of the art of evaluation in neural language models. arXiv preprint arXiv:1707.05589

  39. 39.

    Merity S, Keskar NS, Socher R (2017) Regularizing and optimizing LSTM language models. arXiv preprint arXiv:1708.02182

  40. 40.

    Merity S, Xiong C, Bradbury J, Socher R (2016) Pointer sentinel mixture models. arXiv preprint arXiv:1609.07843

  41. 41.

    Mikolov T, Karafiát M, Burget L, Černockỳ J Khudanpur S (2010) Recurrent neural network based language model. In: Eleventh annual conference of the international speech communication association

  42. 42.

    Mikolov T, Sutskever I, Chen K, Corrado GS, Dean J (2013) Distributed representations of words and phrases and their compositionality. In: Advances in neural information processing systems, pp 3111–3119

  43. 43.

    Molina-González MD, Plaza-del Arco FM, Martín-Valdivia M, Ureña López L (2019) Ensemble learning to detect aggressiveness in mexican spanish tweets. In: Proceedings of the first workshop for Iberian languages evaluation forum (IberLEF 2019), CEUR WS proceedings

  44. 44.

    Pitsilis GK, Ramampiaro H, Langseth H (2018) Detecting offensive language in tweets using deep learning. arXiv preprint arXiv:1801.04433

  45. 45.

    Radford A, Narasimhan K, Salimans T, Sutskever I (2018) Improving language understanding with unsupervised learning. Technical report, OpenAI

  46. 46.

    Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagating errors. Nature 323(6088):533

    Article  Google Scholar 

  47. 47.

    Sharif Razavian A, Azizpour H, Sullivan J, Carlsson S (2014) CNN features off-the-shelf: an astounding baseline for recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition workshops, pp 806–813

  48. 48.

    Silva L, Mondal M, Correa D, Benevenuto F, Weber I (2016) Analyzing the targets of hate in online social media. In: Tenth International AAAI conference on web and social media

  49. 49.

    Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958

    MathSciNet  MATH  Google Scholar 

  50. 50.

    Sun C, Shrivastava A, Singh S, Gupta A (2017) Revisiting unreasonable effectiveness of data in deep learning era. In: Proceedings of the IEEE international conference on computer vision, pp 843–852

  51. 51.

    Sundermeyer M, Schlüter R, Ney H (2012) LSTM neural networks for language modeling. In: Thirteenth annual conference of the international speech communication association

  52. 52.

    Sutskever I, Vinyals O, Le QV (2014) Sequence to sequence learning with neural networks. In: Advances in neural information processing systems, pp 3104–3112

  53. 53.

    Suzor N, Van Geelen T, Myers West S (2018) Evaluating the legitimacy of platform governance: a review of research and a shared research agenda. Int Commun Gaz 80(4):385–400

    Article  Google Scholar 

  54. 54.

    The Online Hate Index: Innovation Brief (2018) Technical report, the anti-defamation League’s center for technology and society.

  55. 55.

    Turian J, Ratinov L, Bengio Y (2010), Word representations: a simple and general method for semi-supervised learning. In: Proceedings of the 48th annual meeting of the association for computational linguistics. Association for Computational Linguistics, pp 384–394

  56. 56.

    Wang B, Wang A, Chen F, Wang Y, Kuo C-CJ (2019) Evaluating word embedding models: methods and experimental results. APSIPA Trans Signal Inf Process 8:e19

    Article  Google Scholar 

  57. 57.

    Wang D, Nyberg E (2015) A long short-term memory model for answer sentence selection in question answering. In: Proceedings of the 53rd annual meeting of the association for computational linguistics and the 7th international joint conference on natural language processing (volume 2: short papers), vol 2, pp 707–712

  58. 58.

    Wang W, Chen L, Thirunarayan K, Sheth AP (2014) Cursing in english on twitter. In: Proceedings of the 17th ACM conference on computer supported cooperative work & social computing. ACM, pp 415–425

  59. 59.

    Weinberger KQ, Saul LK (2009) Distance metric learning for large margin nearest neighbor classification. J Mach Learn Res 10(Feb):207–244

    MATH  Google Scholar 

  60. 60.

    Xiang G, Fan B, Wang L, Hong J, Rose C (2012) Detecting offensive tweets via topical feature discovery over a large scale twitter corpus. In: Proceedings of the 21st ACM international conference on Information and knowledge management. ACM, pp 1980–1984

  61. 61.

    Zeiler MD, Fergus R (2014) Visualizing and understanding convolutional networks. In: European conference on computer vision. Springer, pp 818–833

  62. 62.

    Zhang KW, Bowman SR (2018) Language modeling teaches you more syntax than translation does: lessons learned through auxiliary task analysis. arXiv preprint arXiv:1809.10040

  63. 63.

    Zhang Z, Luo L (2019) Hate speech detection: a solved problem? The challenging case of long tail on twitter. Semant Web 10(5):925–945

    Article  Google Scholar 

Download references


his research was partially supported by the QUT IFE Catapult fund. Suzor is the recipient of an Australian Research Council DECRA Fellowship (project number DE160101542).

Author information



Corresponding author

Correspondence to Md Abul Bashar.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix A: Description of evaluation measures

Appendix A: Description of evaluation measures

  • True Positive (TP) True positives are instances classified as positive by the model that actually are positive.

  • True Negative (TN): True negatives are instances the model classifies as negative that actually are negative.

  • False Positive (FP): False positives are instances identified by model as positive that actually are negative.

  • False Negative (FN): False negatives are instances the model classifies as negative that actually are positive.

  • Accuracy (Ac): It is the percentage of correctly classified instances, and it is calculated as \(\frac{\hbox {TP} + \hbox {TN}}{\hbox {TP} + \hbox {TN} + \hbox {FP} + \hbox {FN}}\).

  • Precision (Pr): It calculates a model’s ability to return only relevant instances. It is calculated as \(\frac{\hbox {TP}}{\hbox {TP} + \hbox {FP}}\).

  • Recall (Re): It calculates a model’s ability to identify all relevant instances. It is calculated as \(\frac{\hbox {TP}}{\hbox {TP} + \hbox {FN}}\).

  • \(F_1\) Score (\(F_1\)): A single metric that combines recall and precision using the harmonic mean. \(F_1\) Score is calculated as \(2 \times \frac{\hbox {precision}}{\hbox {precision} + \hbox {recall}}\).

  • Cohen Kappa (CK): Cohen’s kappa score is used to measure inter-rater and intra-rater reliability for categorical items [37]. It is calculated as \(\frac{\hbox {OA}-\hbox {AC}}{1-\hbox {AC}}\), where OA is the relative observed agreement between predicted labels and actual labels and AC is the probability of agreement by chance.

  • Area Under Curve (AUC): Area under the receiver operating characteristic (ROC) curve is called area under the curve (AUC). ROC plots the true positive rate versus the false positive rate as a function of the model’s threshold for classifying a positive. AUC calculates the overall performance of a classification model.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Bashar, M.A., Nayak, R. & Suzor, N. Regularising LSTM classifier by transfer learning for detecting misogynistic tweets with small training set. Knowl Inf Syst 62, 4029–4054 (2020).

Download citation


  • Misogynistic tweet
  • Transfer learning
  • LSTM
  • Small dataset
  • Overfitting