AI-DRIVEN CROSS-PLATFORM APPLICATION DEVELOPMENT SYSTEMS

Сергій Юрійович Манаков, Олена Григорівна Трофименко, Павло Олександрович Чикунов, Володимир Ігорович Гура

Abstract


The study is devoted to the analysis of the current state and prospects for the development of artificial intelligence (AI) in cross-platform application development systems. The key machine learning technologies used in software engineering automation processes are considered, including automatic code generation systems based on large language models, intelligent development environments, and AI-driven testing methodologies. The architectural solutions of modern cross-platform frameworks and their integration with artificial intelligence technologies are analyzed. The application of transformer models, in particular GPT-4/Codex, Claude, CodeT5, and CodeBERT, in the tasks of understanding and generating software code is studied. The analysis showed that currently GPT-4/Codex is the most accurate and powerful AI model suitable for complex code generation. CodeT5 maintains a balance between size and performance and therefore is well suited for code transformation tasks. InCoder specializes in filling in code patterns but has lower accuracy. CodeBERT is more suitable for code analytics than for generation. Methods for assessing the quality of AI-generated code are considered, including metrics of functional correctness and structural quality. Challenges to the security and reliability of automatically generated software code are highlighted, including vulnerability issues and the need for additional verification. An analysis of the effectiveness of various approaches to cross-platform development using artificial intelligence tools is presented. The scientific novelty of the work lies in the comprehensive analysis of the interaction of artificial intelligence technologies with cross-platform development frameworks, the systematization of modern approaches to AI-driven code generation, and the study of specific challenges in integrating machine learning into multi-platform software development processes. The results of the study show the significant potential of integrating artificial intelligence technologies to increase developer productivity and improve software quality.

Keywords


cross-platform development, artificial intelligence, code generation, large language models, software engineering, testing, integrated development environment

References


AI Trends Report 2024: AI’s Growing Role in Software Development. URL: https://www.docker.com/blog/ai-trends-report-2024/

Wang Y., Wang W., Joty S., Hoi St. CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 2021. P. 8696-8708. DOI: https://doi.org/10.18653/v1/2021.emnlp-main.685

Lu S., Guo D., Ren S., et al. CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation. arXiv. 2021. Vol. 2102.04664. DOI: https://doi.org/10.48550/arXiv.2102.04664

Feng Z., Guo D., Tang D., et al. CodeBERT: A Pre-Trained Model for Programming and Natural Languages. Findings of the Association for Computational Linguistics: EMNLP 2020. P. 1536-1547. DOI: https://doi.org/10.18653/v1/2020.findings-emnlp.139

Fried D., Aghajanyan A., Lin J., et al. InCoder: A Generative Model for Code Infilling and Synthesis. arXiv. 2022. Vol. 2204.05999. DOI: https://doi.org/10.48550/arXiv.2204.05999

Negri-Ribalta C., Geraud-Stewart R., Sergeeva A., Lenzini G. A systematic literature review on the impact of AI models on the security of code generation. Frontiers in Big Data. 2024. Vol. 7. Article 1386720. P. 1-20. DOI: https://doi.org/10.3389/fdata.2024.1386720

Martinović B., Rozić R. Impact of AI Tools on Software Development Code Quality. Communications in Computer and Information Science. 2024. Vol. 2124. P. 202-216. DOI: https://doi.org/10.1007/978-3-031-62058-4_15

Ukrainian Catholic University Machine Learning Lab. Research Portfolio 2020-2025. Lviv: UCU, 2025. URL: https://apps.ucu.edu.ua/en/mllab/

Ma L., Wang H., Yang K, et al. An empirical study of AI techniques in mobile applications. Journal of Systems and Software. 2024. Vol. 220. Article 112233. DOI: https://doi.org/10.1016/j.jss.2024.112233

Blanco J. Z., Lucrédio D. A holistic approach for cross-platform software development. Journal of Computer Languages. 2021. Vol. 65. Article 101049. DOI: https://doi.org/10.48550/arXiv.2104.14614

Alenezi M., Akour M. AI-Driven Innovations in Software Engineering: A Review of Current Practices and Future Directions. Applied Sciences. 2025. Vol. 15, No. 3. Article 1344. DOI: https://doi.org/10.3390/app15031344

Peng S., Kalliamvakou E., Cihon P., Demirer M. The Impact of AI on Developer Productivity: Evidence from GitHub Copilot. arXiv. 2023. Vol. 2302.06590. DOI: https://doi.org/10.48550/arXiv.2302.06590

Chen M., Tworek J., Jun H., et al. Evaluating Large Language Models Trained on Code. arXiv. 2021. Vol. 2107.03374. DOI: https://doi.org/10.48550/arXiv.2107.03374

Fu Ju., Liang P., Tahir A., et al. Security Weaknesses of Copilot-Generated Code in GitHub Projects: An Empirical Study. arXiv. 2025. Vol. 2310:02059v4. DOI: https://doi.org/10.48550/arXiv.2310.02059

CodeScene ACE: Auto-Refactor Code. Technical Documentation. 2025. URL: https://codescene.io/docs/auto-refactor/index.html#codescene-ace-auto-refactor-code

Maltseva A. JetBrains AI Assistant: Smarter, More Capable, and a New Free Tier. 2025. URL: https://blog.jetbrains.com/ai/2025/04/jetbrains-ai-assistant-2025-1/

Трофименко О. Г., Дика А. І., Лобода Ю. Г. Аналіз інструментів тестування вебзастосунків. Кібербезпека: освіта, наука, техніка. 2023. № 4(20). С. 62-71. DOI: https://doi.org/10.28925/2663-4023.2023.20.6271

Ramchand S., Shaikh S., Alam I. Role of Artificial Intelligence in Software Quality Assurance. Lecture Notes in Networks and Systems. 2022. Vol. 295. P. 89 102. DOI: https://doi.org/10.1007/978-3-030-82196-8_10

Ren S., Guo D., Lu S., et al. CodeBLEU: a Method for Automatic Evaluation of Code Synthesis. arXiv. 2020. Vol. 2009.10297. DOI: https://doi.org/10.48550/arXiv.2009.10297

Bhatt M. et al. CyberSecEval 2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models. arXiv. 2024. Vol. 2404.13161. DOI: https://doi.org/10.48550/arXiv.2404.13161

Трофименко, О. Г., Лобода, Ю. Г., Гура, В. І., Дика, А. І., Стрілець, М. І. Інструменти штучного інтелекту для системного аналізу. Вісник Херсонського національного технічного університету. 2024. № 4. С. 349–357. DOI: https://doi.org/10.35546/kntu2078-4481.2024.4.46




DOI: https://doi.org/10.32620/oikit.2025.105.15

Refbacks

  • There are currently no refbacks.