Survey of Machine Learning Classification Techniques for Software Defects Prediction
The Asian Bulletin of Big Data Management · 30.03.2026 · 6 (1) · 293-316
DOI: 10.62019/jeg1vv20 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Asian Bulletin of Big Data Management, online erschienen am 30.03.2026. Verfasst von Quratulain. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
This review paper critically assesses the range of machine learning (ML) approaches used in software defect prediction, with an emphasis on advances in Mathematical Modeling and Simulation. Software defect prediction (SDP) continues to be an important component of software project planning, having a major influence on time to delivery estimates, maintenance activities, and overall performance standards. Before software deployment, some defects need to be forecasted by organizations so during this phase the challenges that occur are investigated in the study to emphasize SDP significance in domains such as robotics, healthcare, aviation, and manufacturing. This review explores the SDP models in terms of evolution specifically highlighting ML classifier integration with static program metrics. Recognizing the possible limits of human feature engineering in missing vital information, the research examines previous SDP outcomes and provides adaptive strategies for future systems to identify anomalies. Machine learning approaches that include Random forest, Multi-layer Perceptron (MLP), K-Means, and support vector machines are reviewed. Further SDP model's current state is discussed and their effect on software quality is improved in the phase of maintenance.SDP models include the evaluation of metrics such as detection rate, false alarm rate, and precision. The review findings highlight the high accuracy achieved by K-Means and SVM (99.19%), K-Means and RF (97.76%), and K-Means and MLP (99.67%) in predicting defects and provide insights into ML technique's effectiveness about SDP.
Zitieren
APA
Quratulain (2026). Survey of Machine Learning Classification Techniques for Software Defects Prediction. The Asian Bulletin of Big Data Management, 6(1), 293-316. https://doi.org/10.62019/jeg1vv20
Vancouver
Quratulain. Survey of Machine Learning Classification Techniques for Software Defects Prediction. The Asian Bulletin of Big Data Management. 2026;6(1):293-316. doi:10.62019/jeg1vv20
BibTeX
@article{Quratulain2026,
author = {Quratulain},
title = {Survey of Machine Learning Classification Techniques for Software Defects Prediction},
journal = {The Asian Bulletin of Big Data Management},
year = {2026},
volume = {6},
number = {1},
pages = {293--316},
doi = {10.62019/jeg1vv20}
}
EUVIMED