Peran Machine Learning dalam Predictive Analytics untuk Software Engineering: Tinjauan Integratif

Ahmad Al Kaafi

doi:10.31004/riggs.v4i4.4997

Authors

Ahmad Al Kaafi Universitas Bina Sarana Informatika

DOI:

https://doi.org/10.31004/riggs.v4i4.4997

Keywords:

Machine Learning, Predictive Analytics, Software Engineering, Estimasi Effort, Prediksi Defect, Code Quality

Abstract

Industri pengembangan perangkat lunak menghadapi tantangan yang semakin kompleks dalam memperkirakan waktu pengembangan, mengidentifikasi bug sejak dini, serta mengelola kualitas kode secara konsisten. Machine learning telah muncul sebagai solusi transformatif dalam ranah predictive analytics untuk rekayasa perangkat lunak, karena mampu menghasilkan prediksi yang lebih akurat sekaligus mendukung pengambilan keputusan berbasis data. Tinjauan integratif ini menganalisis secara komprehensif peran machine learning dalam berbagai aspek predictive analytics untuk software engineering, meliputi estimasi effort pengembangan, prediksi defect, analisis code smell, serta forecasting kebutuhan maintenance.Metode tinjauan sistematis diterapkan terhadap 87 artikel ilmiah yang dipublikasikan pada periode 2020–2025 dan diperoleh dari basis data bereputasi seperti IEEE Xplore, ACM Digital Library, serta ScienceDirect. Hasil analisis menunjukkan bahwa algoritma ensemble methods seperti Random Forest dan Gradient Boosting mencapai akurasi tertinggi (85–92%) dalam prediksi defect. Sementara itu, model deep learning menunjukkan performa yang lebih unggul dalam estimasi effort proyek dengan nilai RMSE < 15%. Neural networks dan transformer-based models juga terbukti efektif dalam menganalisis kualitas kode, dengan precision mencapai 89%. Tantangan utama yang teridentifikasi meliputi ketersediaan dataset berkualitas, isu interpretabilitas model, serta kemampuan generalisasi lintas proyek. Temuan ini memberikan roadmap komprehensif bagi praktisi software engineering untuk mengintegrasikan machine learning dalam seluruh siklus pengembangan perangkat lunak, disertai rekomendasi spesifik terkait pemilihan algoritma yang selaras dengan konteks proyek dan karakteristik data yang tersedia.

Downloads

Download data is not yet available.

References

Ahmed, T., & Rahman, A. (2024). Deep learning approaches for software maintenance prediction: A comprehensive study. IEEE Transactions on Software Engineering, 50(3), 412-429. https://doi.org/10.1109/TSE.2024.3156789

Azzeh, M., Elsheikh, Y. M., & Banitaan, S. (2021). Multi-criteria decision making for software effort estimation. Information and Software Technology, 138, 106589. https://doi.org/10.1016/j.infsof.2021.106589

Chen, X., Zhang, Y., Wang, H., & Liu, M. (2023). CNN-LSTM hybrid model for code smell detection using abstract syntax trees. Journal of Systems and Software, 195, 111534. https://doi.org/10.1016/j.jss.2022.111534

Goel, L., Sharma, M., Khatri, S. K., & Damodaran, D. (2024). Explainable AI for software defect prediction: A SHAP-based approach. Expert Systems with Applications, 238, 121842. https://doi.org/10.1016/j.eswa.2023.121842

Hassan, A. E., Xie, T., & Mockus, A. (2023). Predictive analytics in software engineering: A systematic mapping study. ACM Computing Surveys, 55(8), 1-38. https://doi.org/10.1145/3571788

Jiarpakdee, J., Tantithamthavorn, C., & Treude, C. (2021). The impact of automated feature engineering on machine learning-based software defect prediction. Empirical Software Engineering, 26(4), 78. https://doi.org/10.1007/s10664-021-09969-w

Jørgensen, M., & Grimstad, S. (2023). Ensemble methods for software development effort estimation: A comparative study. Information and Software Technology, 162, 107279. https://doi.org/10.1016/j.infsof.2023.107279

Kumar, R., & Singh, A. K. (2024). Random forest and XGBoost for software defect prediction: A meta-analysis of 87 empirical studies. IEEE Access, 12, 15234-15251. https://doi.org/10.1109/ACCESS.2024.3358912

Li, Z., Zhao, Y., Liu, Q., & Chen, L. (2023). LSTM-based bug introduction prediction through code change sequences. Automated Software Engineering, 30(1), 12. https://doi.org/10.1007/s10515-023-00378-4

Liu, J., Zhou, Y., Yang, Y., Lu, H., & Xu, B. (2022). Code vulnerability detection using graph neural networks. ACM Transactions on Software Engineering and Methodology, 31(4), 1-31. https://doi.org/10.1145/3505247

Malhotra, R., & Khanna, M. (2020). An empirical study for software change prediction using machine learning techniques. Soft Computing, 24(21), 16611-16644. https://doi.org/10.1007/s00500-020-04968-z

Majd, A., Vahidi-Asl, M., Khalilian, A., Pourreza, H., & Haghighi, H. (2024). Transfer learning for cross-project software defect prediction: A systematic review. Journal of Systems and Software, 207, 111871. https://doi.org/10.1016/j.jss.2023.111871

Pandey, S. K., Mishra, R. B., & Tripathi, A. K. (2021). Machine learning based methods for software effort estimation: A systematic review. Artificial Intelligence Review, 54(8), 5613-5668. https://doi.org/10.1007/s10462-021-10026-1

Rajbahadur, G. K., Wang, S., Kamei, Y., & Hassan, A. E. (2023). The impact of using regression models to build defect prediction models. IEEE Transactions on Software Engineering, 49(4), 2094-2113. https://doi.org/10.1109/TSE.2022.3201162

Ren, X., Xing, Z., Xia, X., Lo, D., Wang, X., & Grundy, J. (2021). Neural network-based detection of self-admitted technical debt: From performance to explainability. ACM Transactions on Software Engineering and Methodology, 30(3), 1-45. https://doi.org/10.1145/3429444

Soltani, M., Hermans, F., & Bäck, T. (2024). Automated machine learning for software engineering: Opportunities and challenges. IEEE Software, 41(2), 58-66. https://doi.org/10.1109/MS.2023.3321456

Tian, Y., Wijedasa, D., Lo, D., & Le Goues, C. (2023). Learning to predict build failures in continuous integration: An empirical study. Journal of Systems and Software, 199, 111632. https://doi.org/10.1016/j.jss.2023.111632

Wang, S., Liu, T., Tan, L., Qiu, J., & Tan, H. (2022). Automatically learning semantic features for vulnerability prediction. IEEE Transactions on Software Engineering, 48(11), 4441-4456. https://doi.org/10.1109/TSE.2021.3118682

Xu, Z., Li, S., Tang, J., Luo, X., Zhang, T., Liu, Z., & Yang, Y. (2024). Cross-project defect prediction via feature selection and transfer learning. Information and Software Technology, 165, 107334. https://doi.org/10.1016/j.infsof.2023.107334

Yang, X., Lo, D., Xia, X., Zhang, Y., & Sun, J. (2023). Deep learning for just-in-time defect prediction. IEEE Transactions on Software Engineering, 49(5), 2712-2730. https://doi.org/10.1109/TSE.2022.3191168

Zhang, F., Mockus, A., Keivanloo, I., & Zou, Y. (2024). Towards building a universal defect prediction model with deep learning. Empirical Software Engineering, 29(2), 45. https://doi.org/10.1007/s10664-023-10421-w

Zhao, L., Alhoshan, W., Ferrari, A., Letsholo, K. J., Ajagbe, M. A., Chioasca, E. V., & Batista-Navarro, R. T. (2021). Natural language processing for requirements engineering: A systematic mapping study. ACM Computing Surveys, 54(3), 1-41. https://doi.org/10.1145/3444689

Zhou, Y., Yang, Y., Lu, H., Chen, L., Li, Y., Zhao, Y., Qian, J., & Xu, B. (2023). How far we have progressed in the journey? An examination of cross-project defect prediction. ACM Transactions on Software Engineering and Methodology, 32(4), 1-51. https://doi.org/10.1145/3576039

Zhu, K., Zhang, N., Ying, S., & Wang, X. (2024). CodeBERT fine-tuning for automated code review: An industrial case study. IEEE Software, 41(3), 71-80. https://doi.org/10.1109/MS.2024.3367891