Artificial intelligence

Abstract: This paper presents an approach to analyze the web applications test coverage using machine learning methods. The research explores neural embedding techniques for web application state abstraction, classification algorithms for identifying uncovered execution paths, and reinforcement learning agents for automated UI exploration. The study examines automatic test case generation methods for uncovered code regions, including large language model (LLM) approaches such as TestPilot and AUTOE2E, mutation-guided testing with MuTAP, and hybrid GAN+LLM architectures. Test suite optimization techniques are analyzed, including duplicate removal through embeddings, test prioritization, and self-healing mechanisms. The paper proposes an integrated six-level architecture for an intelligent test coverage analysis system. Experimental results on a benchmark of 15 web applications show the proposed approach detects 35% more defects compared to traditional methods.

Keywords: test coverage analysis, machine learning, web application testing, large language models, neural embeddings, automated test generation, mutation testing

References:

1. Chang, X., Chen, Y., Wang, Z., & Zhang, L. (2023). A Reinforcement Learning Approach to Generating Test Cases for Web Applications. 2023 IEEE/ACM International Conference on Automation of Software Test (AST), 1-11. https://doi.org/10.1109/AST58925.2023.00006
2. Ale, N. K. (2024). A Generative AI Framework for Enhancing Software Test Automation. International Journal of Science and Research (IJSR), 13(6), 604-612. https://doi.org/10.21275/sr24604032016
3. Emektar, M., Yilmaz, C., & Ozturk, K. (2025). AI-Driven Synthetic Test Data and Scenario Generation via GAN-LLM Integration. 2025 Innovations in Intelligent Systems and Applications Conference (ASYU), 1-6. https://doi.org/10.1109/ubmk67458.2025.11206764
4. Sharma, R., & Kumar, A. (2023). Software Engineering and Automation: Emerging Trends and Challenges. International Conference on Computer Science and Information Technology, 13(20), 145-158. https://doi.org/10.5121/csit.2023.1320
5. Junior, E. S., Silva, M. A., & Santos, R. P. (2025). GenIA-E2ETest: A Generative AI-Based Approach for End-to-End Test Automation. Anais do XXXIX Simpósio Brasileiro de Engenharia de Software, 123-134. https://doi.org/10.5753/sbes.2025.9927
6. Tao, L., Wang, H., Chen, J., & Liu, Y. (2024). A Survey on Web Application Testing: A Decade of Evolution. arXiv preprint arXiv:2412.10476. https://doi.org/10.48550/arxiv.2412.10476
7. Patel, S., & Gupta, M. (2024). AI and Machine Learning in Automated Test Case Generation: Emerging Trends. Neliti Publications. https://www.neliti.com/publications/603898/
8. Khaliq, Z., Rana, R., Ahsan, M., & Lali, M. I. (2022). A deep learning-based automated framework for functional UI testing. Information and Software Technology, 150, 106969. https://doi.org/10.1016/j.infsof.2022.106969
9. Leotta, M., Stocco, A., Ricca, F., & Tonella, P. (2024). An empirical study to compare three web test automation approaches. Software Testing, Verification and Reliability, 34(3), e2606. https://doi.org/10.1002/smr.2606
10. Alian, P., Behnamghader, P., & Halfond, W. G. J. (2024). A Feature-Based Approach to Generating Comprehensive End-to-End Tests. arXiv preprint arXiv:2408.01894. https://doi.org/10.48550/arxiv.2408.01894
11. Celik, A., & Mahmoud, A. (2025). Large Language Models for Test Case Generation: A Systematic Review. Machine Learning and Knowledge Extraction, 7(3), 1847-1872. https://doi.org/10.3390/make7030097
12. Deng, Y., Xia, C. S., Peng, H., Yang, C., & Zhang, L. (2023). Large Language Models are Zero-Shot Fuzzers. Proceedings of the 32nd ACM SIGSOFT ISSTA, 423-435. https://doi.org/10.1145/3597926.3598067
13. Schafer, M., Nadi, S., Eghbali, A., & Tip, F. (2023). An Empirical Evaluation of Using Large Language Models for Automated Unit Test Generation. IEEE Transactions on Software Engineering, 49(4), 1411-1427. https://doi.org/10.1109/TSE.2023.3234726
14. Pan, R., et al. (2024). Understanding the Effectiveness of Large Language Models in Detecting Security Vulnerabilities. arXiv:2311.16169. https://doi.org/10.48550/arxiv.2311.16169
15. Siddiq, M. L., & Santos, J. C. S. (2023). SecurityEval Dataset: Mining Vulnerability Examples. Proceedings of MSR 2023, 273-277. https://doi.org/10.1109/MSR59073.2023.00050
16. Prenner, J. A., & Robbes, R. (2021). Automatic Program Repair with OpenAI's Codex: Evaluating QuixBugs. arXiv:2111.03922. https://doi.org/10.48550/arxiv.2111.03922
17. Lemieux, C., Inala, J. P., Lahiri, S. K., & Sen, S. (2023). Codamosa: Escaping Coverage Plateaus in Test Generation with Pre-trained Large Language Models. Proceedings of the 45th ICSE, 919-931. https://doi.org/10.1109/ICSE48619.2023.00085
18. Dinella, E., Ryan, G., Mytkowicz, T., & Lahiri, S. K. (2023). TOGA: A Neural Method for Test Oracle Generation. Proceedings of the 45th ICSE, 1039-1051. https://doi.org/10.1109/ICSE48619.2023.00095
19. Wang, J., Huang, Y., Chen, C., Liu, Z., Wang, S., & Wang, Q. (2024). Software Testing with Large Language Models: Survey, Landscape, and Vision. IEEE Transactions on Software Engineering, 50(3), 555-580. https://doi.org/10.1109/TSE.2024.3368208
20. Chen, C., Bansal, C., Wang, S., Zimmermann, T., & Huang, Y. (2023). AdbGPT: Debugging Automation via Large Language Models. arXiv:2306.04138. https://doi.org/10.48550/arxiv.2306.04138