Exploring model transfer strategies for sentiment analysis in Twitter

• Eliseu Guimarães UFF / Marinha do Brasil
• Jonnathan Carvalho IFF
• Aline Paes UFF
• Alexandre Plastino UFF

DOI: https://doi.org/10.5753/eniac.2021.18236

Abstract

Social media have become trendy environments for communication. Because of that, analyze the sentiment that the user expresses in their social media posts is an important research field. However, detecting polarity in such contents is a challenge, partially because the amount of labeled data to train classifiers is scarce in many situations. This paper explores strategies for reusing a model learned from a source dataset to classify instances in a target dataset. The experiments are conducted with 22 tweets sentiment analysis datasets and approaches based on similarity metrics. The results point out that the size of the source training set plays an essential role in the classifiers’ performance when they were applied to the target data.

References

Barbosa, L. and Feng, J. (2010). Robust sentiment detection on twitter from biased and In Proc. of the 23rd Int. Conf. on Computational Linguistics: Posters, noisy data. COLING ’10, page 36–44. ACL.

Bravo-Marquez, F., Frank, E., Mohammad, S. M., and Pfahringer, B. (2016). Determining word-emotion associations from tweets by multi-label classification. In Proc. of the 2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI), pages 536– 539. IEEE.

Carvalho, J. and Plastino, A. (2021). On the evaluation and combination of state-of-the-art features in twitter sentiment analysis. Artificial Intelligence Review, 54.

Dong, L., Wei, F., Tan, C., Tang, D., Zhou, M., and Xu, K. (2014). Adaptive recursive neural network for target-dependent Twitter sentiment classification. In Proc. of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 49–54. ACL.

Guimarães, E., Carvalho, J., Paes, A., and Plastino, A. (2020). Transfer learning for In Anais do VIII twitter sentiment analysis: Choosing an effective source dataset. Symposium on Knowledge Discovery, Mining and Learning, pages 161–168. SBC.

Guo, J., Shah, D., and Barzilay, R. (2018). Multi-source domain adaptation with mixture of experts. In Proc. of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 4694–4703. ACL.

Li, S., Song, S., and Huang, G. (2017). Prediction reweighting for domain adaptation. IEEE Transactions on Neural Networks and Learning Systems, 28(7):1682–1695.

Liu, B. (2020). Sentiment Analysis: Mining Opinions, Sentiments, and Emotions. Studies in Natural Language Processing. Cambridge University Press, 2 edition.

Martínez-Cámara, E., Martín-Valdivia, M., López, L., and Montejo-Ráez, A. (2014). Sentiment analysis in twitter. Natural Language Engineering, 20:1–28.

Mikolov, T., Chen, K., Corrado, G. S., and Dean, J. (2013). Efficient Estimation of Word Representations in Vector Space. CoRR, abs/1301.3781.

Pan, S. J. and Yang, Q. (2010). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10):1345–1359.

Ruder, S. and Plank, B. (2017). Learning to select data for transfer learning with Bayesian Optimization. In Proc. of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 372–382. ACL.

Santos, J. S., Paes, A., and Bernardini, F. (2019). Combining labeled datasets for sentiment analysis from different domains based on dataset similarity to predict electors sentiment. In 2019 8th Brazilian Conf. on Intelligent Systems, pages 455–460.