Predicting Software Perfection Through Advanced Models to Uncover and Prevent Defects

Software defect prediction is a critical task in software engineering, enabling organizations to proactively identify and address potential issues in software systems, thereby improving quality and reducing costs. In this study, we evaluated and compared various machine learning models, including lo...

Full description

Saved in:
Bibliographic Details
Main Authors: Tariq Shahzad, Sunawar Khan, Tehseen Mazhar, Wasim Ahmad, Khmaies Ouahada, Habib Hamam
Format: Article
Language:English
Published: Wiley 2025-01-01
Series:IET Software
Online Access:http://dx.doi.org/10.1049/sfw2/8832164
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850182897384816640
author Tariq Shahzad
Sunawar Khan
Tehseen Mazhar
Wasim Ahmad
Khmaies Ouahada
Habib Hamam
author_facet Tariq Shahzad
Sunawar Khan
Tehseen Mazhar
Wasim Ahmad
Khmaies Ouahada
Habib Hamam
author_sort Tariq Shahzad
collection DOAJ
description Software defect prediction is a critical task in software engineering, enabling organizations to proactively identify and address potential issues in software systems, thereby improving quality and reducing costs. In this study, we evaluated and compared various machine learning models, including logistic regression (LR), random forest (RF), support vector machines (SVMs), convolutional neural networks (CNNs), and eXtreme Gradient Boosting (XGBoost), for software defect prediction using a combination of diverse datasets. The models were trained and tested on preprocessed and feature-selected data, followed by optimization through hyperparameter tuning. Performance evaluation metrics were employed to analyze the results comprehensively, including classification reports, confusion matrices, receiver operating characteristic–area under the curve (ROC-AUC) curves, precision–recall curves, and cumulative gain charts. The results revealed that XGBoost consistently outperformed other models, achieving the highest accuracy, precision, recall, and AUC scores across all metrics. This indicates its robustness and suitability for predicting software defects in real-world applications.
format Article
id doaj-art-e788f199159e4d3e93c74d8dd054da33
institution OA Journals
issn 1751-8814
language English
publishDate 2025-01-01
publisher Wiley
record_format Article
series IET Software
spelling doaj-art-e788f199159e4d3e93c74d8dd054da332025-08-20T02:17:32ZengWileyIET Software1751-88142025-01-01202510.1049/sfw2/8832164Predicting Software Perfection Through Advanced Models to Uncover and Prevent DefectsTariq Shahzad0Sunawar Khan1Tehseen Mazhar2Wasim Ahmad3Khmaies Ouahada4Habib Hamam5Department of Electrical and Electronic Engineering ScienceDepartment of Software EngineeringSchool of Computer ScienceDepartment of ComputingDepartment of Electrical and Electronic Engineering ScienceDepartment of Electrical and Electronic Engineering ScienceSoftware defect prediction is a critical task in software engineering, enabling organizations to proactively identify and address potential issues in software systems, thereby improving quality and reducing costs. In this study, we evaluated and compared various machine learning models, including logistic regression (LR), random forest (RF), support vector machines (SVMs), convolutional neural networks (CNNs), and eXtreme Gradient Boosting (XGBoost), for software defect prediction using a combination of diverse datasets. The models were trained and tested on preprocessed and feature-selected data, followed by optimization through hyperparameter tuning. Performance evaluation metrics were employed to analyze the results comprehensively, including classification reports, confusion matrices, receiver operating characteristic–area under the curve (ROC-AUC) curves, precision–recall curves, and cumulative gain charts. The results revealed that XGBoost consistently outperformed other models, achieving the highest accuracy, precision, recall, and AUC scores across all metrics. This indicates its robustness and suitability for predicting software defects in real-world applications.http://dx.doi.org/10.1049/sfw2/8832164
spellingShingle Tariq Shahzad
Sunawar Khan
Tehseen Mazhar
Wasim Ahmad
Khmaies Ouahada
Habib Hamam
Predicting Software Perfection Through Advanced Models to Uncover and Prevent Defects
IET Software
title Predicting Software Perfection Through Advanced Models to Uncover and Prevent Defects
title_full Predicting Software Perfection Through Advanced Models to Uncover and Prevent Defects
title_fullStr Predicting Software Perfection Through Advanced Models to Uncover and Prevent Defects
title_full_unstemmed Predicting Software Perfection Through Advanced Models to Uncover and Prevent Defects
title_short Predicting Software Perfection Through Advanced Models to Uncover and Prevent Defects
title_sort predicting software perfection through advanced models to uncover and prevent defects
url http://dx.doi.org/10.1049/sfw2/8832164
work_keys_str_mv AT tariqshahzad predictingsoftwareperfectionthroughadvancedmodelstouncoverandpreventdefects
AT sunawarkhan predictingsoftwareperfectionthroughadvancedmodelstouncoverandpreventdefects
AT tehseenmazhar predictingsoftwareperfectionthroughadvancedmodelstouncoverandpreventdefects
AT wasimahmad predictingsoftwareperfectionthroughadvancedmodelstouncoverandpreventdefects
AT khmaiesouahada predictingsoftwareperfectionthroughadvancedmodelstouncoverandpreventdefects
AT habibhamam predictingsoftwareperfectionthroughadvancedmodelstouncoverandpreventdefects