TEXTUAL ADVERSARIAL EXAMPLE GENERATION USING BIGRAM UNIGRAM-SEMANTIC PRESERVATION OPTIMIZATION ALGORITHM
DOI:
https://doi.org/10.24191/mjoc.vo11i1.11057Keywords:
Adversarial Example, NLP, Semantic Similarity, Web Based ApplicationAbstract
The vulnerability of Natural Language Processing (NLP) models to adversarial attacks remains a critical challenge in the field of cybersecurity and AI robustness. While deep learning models have achieved high performance in sentiment analysis, they are susceptible to subtle input perturbations that induce misclassification. This study presents the design and practical implementation of a web-based system (Proof of Concept) that automates the generation of textual adversarial examples using the Bigram Unigram-Semantic Preservation Optimization (BU-SPOF) algorithm. Rather than proposing a novel attack algorithm, our primary contribution is the architectural integration of a dual-source candidate generation strategy (WordNet and OpenHowNet) and a Probability Weighted Word Saliency (PWWS) mechanism to perturb input text while maintaining linguistic coherence. The system was evaluated against a Long Short-Term Memory (LSTM) sentiment classifier using the IMDB dataset.
References
Al-Smadi, M., Hammad, M., Al-Zboon, S. A., Al-Tawalbeh, S., & Wang, Z. (2023). Gated recurrent unit with multilingual universal sentence encoder for Arabic aspect-based sentiment analysis. Knowledge-Based Systems, 107540.
Asraf, H., Yahaya, J. H., & Berhan, P. (2018). Word sense disambiguation using fuzzy semantic-based string similarity model. Malaysian Journal of Computing, 3(2), 13-26.
Bajaj, A., & Vishwakarma, D. K. (2023). HOMOCHAR: A novel adversarial attack framework for exposing the vulnerability of text-based neural sentiment classifiers. Engineering Applications of Artificial Intelligence, 126, 106815.
Basit, A., & Ahmad, N. (2024). Predicting COVID-19 Trends: A Deep Dive Into Time-Dependent SIRSD With Deep-Learning Technique. Malaysian Journal of Computing, 9(2), 1955-1978.
Chang, G., Gao, H., Zhou, Y., & Xiong, H. (2023). TextGuise: Adaptive adversarial example attacks on text classification model. Neurocomputing.
Gao, J., Lanchantin, J., Soffa, M. L., & Qi, Y. (2018). Black-Box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers. 2018 IEEE Security and Privacy Workshops (SPW), 50-56.
Jin, D., Jin, Z., Zhou, J. T., & Szolovits, P. (2020). Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment. Proceedings of the AAAI Conference on Artificial Intelligence.
Li, A., Zhang, F., Li, S., Chen, T., Su, P., & Wang, H. (2023). Efficiently generating sentence-level textual adversarial examples with Seq2seq Stacked Auto-Encoder. Expert Systems With Applications, 213, 119170.
Shi, Z., Ma, Y., & Yu, X. (2021). An Effective and Efficient Method for Word-Level Textual Adversarial Attack. 2021 IEEE Symposium on Computers and Communications (ISCC), 1-6.
Wang, J., Bao, R., Zhang, Z., & Zhao, H. (2022). Rethinking Textual Adversarial Defense for Pre-Trained Language Models. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 30, 2526-2540.
Yang, X., Gong, Y., Liu, W., Bailey, J., Tao, D., & Liu, W. (2023). Semantic-Preserving Adversarial Text Attacks. IEEE Transactions on Sustainable Computing, 8(4), 583-595.
Ye, S., Zhang, P., Dong, H., & Ji, S. (2021). Heuristic-word-selection Genetic Algorithm for Generating Natural Language Adversarial Examples. 2021 IEEE International Conference on Artificial Intelligence Testing (AITest), 39-40.
Yu, X., Yin, Q., Shi, Z., & Ma, Y. (2022). Improving the Semantic Consistency of Textual Adversarial Attacks via Prompt. 2022 International Joint Conference on Neural Networks (IJCNN), 1-8.
Zhang, H., Xie, Y., Zhu, Z., Sun, J., Li, C., & Gu, Z. (2021). Attack-words Guided Sentence Generation for Textual Adversarial Attack. 2021 IEEE Sixth International Conference on Data Science in Cyberspace (DSC), 280-287.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Noor Adam Bin Noor Azmi, Ts. Dr. Haslizatul Fairuz Binti Mohamed Hanum
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
