journal article·2024-05-27Open access

A Survey of Adversarial Attacks: An Open Issue for Deep Learning Sentiment Analysis Models

Monserrat Vázquez-Hernández, Luis Alberto Morales-Rosales, Ignacio Algredo-Badillo, Sofía Isabel Fernández-Gregorio, Héctor Rodríguez-Rangel, María-Luisa Córdoba-Tlaxcalteco

Applied Sciences· Volume 14· Issue 11

Abstract

In recent years, the use of deep learning models for deploying sentiment analysis systems has become a widespread topic due to their processing capacity and superior results on large volumes of information. However, after several years’ research, previous works have demonstrated that deep learning models are vulnerable to strategically modified inputs called adversarial examples. Adversarial examples are generated by performing perturbations on data input that are imperceptible to humans but that can fool deep learning models’ understanding of the inputs and lead to false predictions being generated. In this work, we collect, select, summarize, discuss, and comprehensively analyze research works to generate textual adversarial examples. There are already a number of reviews in the existing literature concerning attacks on deep learning models for text applications; in contrast to previous works, however, we review works mainly oriented to sentiment analysis tasks. Further, we cover the related information concerning generation of adversarial examples to make this work self-contained. Finally, we draw on the reviewed literature to discuss adversarial example design in the context of sentiment analysis tasks.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Adversarial Robustness in Machine Learning · Advanced Malware Detection Techniques · Network Security and Intrusion Detection

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Department of Computer Science, CONACYT-National Institute for Astrophysics, Optics and Electronics, Luis Enrique Erro #1, Sta María Tonanzintla, Puebla 72840, Puebla, Mexico

Facultad de Ingeniería Civil, CONACYT-Universidad Michoacana de San Nicolás de Hidalgo, C. de Santiago Tapia 403, Centro, Morelia 58000, Michoacan, Mexico

Facultad de Estadística e Informática, Univesidad Veracruzana, Av. Xalapa, Obrero Campesina, Xalapa 91020, Veracruz, Mexico

Tecnológico Nacional de México, Instituto Tecnológico de Culiacán, Juan de Dios Batiz No. 310pte., Culiacán 80220, Sinaloa, Mexico

Funding

Mexican National Council of Humanities Science and Technology (CONAHCYT)

References (51)

Show source-provided references
1. Shaba, M. (2021, January 14–16). A Real-Time Sentimental Analysis on E-Commerce Sites in Nigeria Using Machine Learning. Proceedings of the Hybrid Intelligent Systems: 21st International Conference on Hybrid Intelligent Systems (HIS 2021), Online.
2. Liu, B., and Zhang, L. (2012). Mining Text Data, Springer.
3. Liu, B. (2011). Web Data Mining: Exploring Hyperlinks, Contents, and Usage Data, Springer.10.1007/978-3-642-19460-3 ↗
4. Multimodal sentimental analysis for social media applications: A comprehensive review10.1002/widm.1415 ↗
5. Multimodal sentiment analysis based on fusion methods: A survey10.1016/j.inffus.2023.02.028 ↗
6. Deep learning: A comprehensive overview on techniques, taxonomy, applications and research directions
7. Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., and Fergus, R. (2013). Intriguing properties of neural networks. arXiv.
8. Goodfellow, I.J., Shlens, J., and Szegedy, C. (2014). Explaining and Harnessing Adversarial Examples. arXiv.
9. Jia, R., and Liang, P. (2017, January 9–11). Adversarial Examples for Evaluating Reading Comprehension Systems. Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, Copenhagen, Denmark.10.18653/v1/D17-1215 ↗
10. Adversarial attacks on deep-learning models in natural language processing: A survey
11. Meng, D., and Chen, H. (November, January 30). Magnet: A two-pronged defense against adversarial examples. Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security, Dallas, TX, USA.10.1145/3133956.3134057 ↗
12. Alsmadi, I., Ahmad, K., Nazzal, M., Alam, F., Al-Fuqaha, A., Khreishah, A., and Algosaibi, A. (2021). Adversarial attacks and defenses for social network text processing applications: Techniques, challenges and future research directions. arXiv.
13. Adversarial attacks and defenses in images, graphs and text: A review10.1007/s11633-019-1211-x ↗
14. Text Adversarial Attacks and Defenses: Issues, Taxonomy, and Perspectives10.1155/2022/6458488 ↗
15. Representation learning: A review and new perspectives10.1109/TPAMI.2013.50 ↗
16. Gong, Z., Wang, W., Li, B., Song, D., and Ku, W.S. (2018). Adversarial Texts with Gradient Methods. arXiv.
17. Tsai, Y.T., Yang, M.C., and Chen, H.Y. (2019, January 1). Adversarial Attack on Sentiment Classification. Proceedings of the 2019 ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP, Florence, Italy.10.18653/v1/W19-4824 ↗
18. Li, J., Ji, S., Du, T., Li, B., and Wang, T. (2019, January 24–27). TextBugger: Generating Adversarial Text Against Real-world Applications. Proceedings of the 2019 Network and Distributed System Security Symposium, San Diego, CA, USA.10.14722/ndss.2019.23138 ↗
19. Alzantot, M., Sharma, Y., Elgohary, A., Ho, B.J., Srivastava, M., and Chang, K.W. (November, January 31). Generating Natural Language Adversarial Examples. Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium.10.18653/v1/D18-1316 ↗
20. Adversarial examples: Attacks and defenses for deep learning10.1109/TNNLS.2018.2886017 ↗
21. Targeted Adversarial Attack Against Deep Cross-Modal Hashing Retrieval10.1109/TCSVT.2023.3263054 ↗
22. Eger, S., Şahin, G.G., Rücklé, A., Lee, J.U., Schulz, C., Mesgar, M., Swarnkar, K., Simpson, E., and Gurevych, I. (2019, January 2–7). Text Processing Like Humans Do: Visually Attacking and Shielding NLP Systems. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Minneapolis, MN, USA. Long and Short Papers.10.18653/v1/N19-1165 ↗
23. Zang, Y., Qi, F., Yang, C., Liu, Z., Zhang, M., Liu, Q., and Sun, M. (2020, January 5–10). Word-level Textual Adversarial Attacking as Combinatorial Optimization. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Online.10.18653/v1/2020.acl-main.540 ↗
24. Gan, W.C., and Ng, H.T. (2019, January 28). Improving the Robustness of Question Answering Systems to Question Paraphrasing. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy.10.18653/v1/P19-1610 ↗
25. Liang, B., Li, H., Su, M., Bian, P., Li, X., and Shi, W. (2018, January 13–19). Deep Text Classification Can be Fooled. Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, IJCAI-18. International Joint Conferences on Artificial Intelligence Organization, Stockholm, Sweden.10.24963/ijcai.2018/585 ↗
26. Vijayaraghavan, P., and Roy, D. (2020). Machine Learning and Knowledge Discovery in Databases, Springer International Publishing.
27. Adversarial examples: Opportunities and challenges
28. Sutskever, I., Vinyals, O., and Le, Q.V. (2014). Advances in Neural Information Processing Systems, Curran Associates, Inc.
29. Bahdanau, D., Cho, K., and Bengio, Y. (2014). Neural Machine Translation by Jointly Learning to Align and Translate. arXiv.
30. Quintero, Y.C., and Garcia, L.A. (2018, January 26–30). Estudio del análisis de sentimentos basado en espectos. Proceedings of the “IV Conferencia Internacional en Ciencias Computacionales e Informáticas”, Havana, Cuba.
31. Graves, A. (2012). Supervised Sequence Labelling with Recurrent Neural Networks, Springer.10.1007/978-3-642-24797-2 ↗
32. Ma, Y., Peng, H., and Cambria, E. (2018, January 2–7). Targeted aspect-based sentiment analysis via embedding commonsense knowledge into an attentive LSTM. Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, New Orleans, LA, USA.10.1609/aaai.v32i1.12048 ↗
33. Syntactic edge-enhanced graph convolutional networks for aspect-level sentiment classification with interactive attention10.1109/ACCESS.2020.3019277 ↗
34. TextFirewall: Omni-Defending Against Adversarial Texts in Sentiment Classification10.1109/ACCESS.2021.3058278 ↗
35. Ribeiro, M.T., Singh, S., and Guestrin, C. (2018, January 15–20). Semantically Equivalent Adversarial Rules for Debugging NLP models. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Melbourne, Australia. Long Papers.10.18653/v1/P18-1079 ↗
36. Gao, J., Lanchantin, J., Soffa, M.L., and Qi, Y. (2018, January 24). Black-box generation of adversarial text sequences to evade deep learning classifiers. Proceedings of the 2018 IEEE Security and Privacy Workshops (SPW), San Francisco, CA, USA.10.1109/SPW.2018.00016 ↗
37. Jin, D., Jin, Z., Zhou, J.T., and Szolovits, P. (2020, January 7–12). Is bert really robust? A strong baseline for natural language attack on text classification and entailment. In Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA.10.1609/aaai.v34i05.6311 ↗
38. Ekbal, A. (2022, January 20–23). Adversarial Sample Generation for Aspect based Sentiment Classification. Proceedings of the Findings of the Association for Computational Linguistics: AACL-IJCNLP 2022, Online.
39. Xu, Y., Zhong, X., Jimeno Yepes, A., and Lau, J.H. (2021, January 6–11). Grey-box Adversarial Attack and Defence For Sentiment Classification. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Online.10.18653/v1/2021.naacl-main.321 ↗
40. Wang, X., Jin, H., Yang, Y., and He, K. (2021, January 27–30). Natural Language Adversarial Defense through Synonym Encoding. Proceedings of the Conference on Uncertainty in Artificial Intelligence, Online.
41. Wang, Z., and Wang, H. (2020, January 28–30). Defense of word-level adversarial attacks via random substitution encoding. Proceedings of the International Conference on Knowledge Science, Engineering and Management, Hangzhou, China.10.1007/978-3-030-55393-7_28 ↗
42. Papernot, N., McDaniel, P., Wu, X., Jha, S., and Swami, A. (2016, January 22–26). Distillation as a defense to adversarial perturbations against deep neural networks. Proceedings of the 2016 IEEE Symposium on Security and Privacy (SP), San Jose, CA, USA.10.1109/SP.2016.41 ↗
43. A survey of adversarial defenses and robustness in nlp10.1145/3593042 ↗
44. Zhou, Y., Zheng, X., Hsieh, C.J., Chang, K.w., and Huang, X. (2020). Defense against adversarial attacks in nlp via dirichlet neighborhood ensemble. arXiv.10.18653/v1/2021.acl-long.426 ↗
45. Towards deep learning models resistant to adversarial attacks
46. Grosse, K., Papernot, N., Manoharan, P., Backes, M., and McDaniel, P. (2017, January 11–15). Adversarial examples for malware detection. Proceedings of the Computer Security—ESORICS 2017: 22nd European Symposium on Research in Computer Security, Oslo, Norway. Part II 22.10.1007/978-3-319-66399-9_4 ↗
47. Pruthi, D., Dhingra, B., and Lipton, Z.C. (August, January 28). Combating Adversarial Misspellings with Robust Word Recognition. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy.
48. Zhou, Y., Jiang, J.Y., Chang, K.W., and Wang, W. (2019). Learning to Discriminate Perturbations for Blocking Adversarial Attacks in Text Classification. arXiv.10.18653/v1/D19-1496 ↗
49. Du, X., Yi, Z., Li, S., Ma, J., Yu, J., Tan, Y., and Wu, Q. (2020, January 19–23). Generating More Effective and Imperceptible Adversarial Text Examples for Sentiment Classification. Proceedings of the International Conference on Artificial Intelligence and Security, Dublin, Ireland.10.1007/978-3-030-57884-8_37 ↗
50. Disrupting adversarial transferability in deep neural networks10.1016/j.patter.2022.100472 ↗
51. Yuan, L., Zheng, X., Zhou, Y., Hsieh, C.J., and Chang, K.W. (2021, January 7–11). On the Transferability of Adversarial Attacks against Neural Text Classifier. Proceedings of the Empirical Methods in Natural Language Processing, Punta Cana, Dominican Republic.10.18653/v1/2021.emnlp-main.121 ↗