Adaptive Ensemble Learning Model-Based Binary White Shark Optimizer for Software Defect Classification
Abstract Software dominates modern enterprises, affecting numerous functions. Software firms constantly experiment with new methodologies to define and assess software quality to stay competitive and ensure excellence. Software engineering uses fundamentals and cutting-edge technology to develop gre...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-01-01
|
| Series: | International Journal of Computational Intelligence Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s44196-024-00716-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585462333571072 |
|---|---|
| author | Jameel Saraireh Mary Agoyi Sofian Kassaymeh |
| author_facet | Jameel Saraireh Mary Agoyi Sofian Kassaymeh |
| author_sort | Jameel Saraireh |
| collection | DOAJ |
| description | Abstract Software dominates modern enterprises, affecting numerous functions. Software firms constantly experiment with new methodologies to define and assess software quality to stay competitive and ensure excellence. Software engineering uses fundamentals and cutting-edge technology to develop great software. In recent decades, Data-mining techniques and machine learning for classifying problematic software projects have emerged to improve software quality. ML approaches, especially ensemble learning models, are becoming fundamental to software engineers’ daily jobs. This work created a binary white shark optimizer (WSO) to optimize standard ensemble learning models. The objective is to identify the most suitable ensemble number for weak learners to maximize accuracy on benchmark datasets. The EM model uses 14 weak learners. Twenty-one experimental runs are performed on 15 software-defective module datasets. The optimized ensemble model outperforms the standard Ensemble learning model in AUC-ROC, Accuracy, Precision, Recall, F1-Score, and Specificity. The enhanced model has an average accuracy of 86%, compared to 76% for the standard ensemble model across all datasets. The optimized model outperformed the conventional ensemble for the same datasets, with an average AUC of 72% compared to 61% for the standard ensemble. The optimized model was more stable than the standard model, with an STD of 5.53E−03 vs 7.24E−02 for the ensemble model. The WSO optimization process strengthens and generalizes optimizeels. The study suggests that evolutionary metaheuristic approaches can enhance EM models’ accuracy, trustworthiness, and adaptability. |
| format | Article |
| id | doaj-art-e80a14b09b6e499ebd78754bf4492c53 |
| institution | Kabale University |
| issn | 1875-6883 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Springer |
| record_format | Article |
| series | International Journal of Computational Intelligence Systems |
| spelling | doaj-art-e80a14b09b6e499ebd78754bf4492c532025-01-26T12:51:41ZengSpringerInternational Journal of Computational Intelligence Systems1875-68832025-01-0118115110.1007/s44196-024-00716-0Adaptive Ensemble Learning Model-Based Binary White Shark Optimizer for Software Defect ClassificationJameel Saraireh0Mary Agoyi1Sofian Kassaymeh2Department of Management Information Systems, Cyprus International UniversityDepartment of Information Technology, Cyprus International UniversityJadara University Research Center, Jadara UniversityAbstract Software dominates modern enterprises, affecting numerous functions. Software firms constantly experiment with new methodologies to define and assess software quality to stay competitive and ensure excellence. Software engineering uses fundamentals and cutting-edge technology to develop great software. In recent decades, Data-mining techniques and machine learning for classifying problematic software projects have emerged to improve software quality. ML approaches, especially ensemble learning models, are becoming fundamental to software engineers’ daily jobs. This work created a binary white shark optimizer (WSO) to optimize standard ensemble learning models. The objective is to identify the most suitable ensemble number for weak learners to maximize accuracy on benchmark datasets. The EM model uses 14 weak learners. Twenty-one experimental runs are performed on 15 software-defective module datasets. The optimized ensemble model outperforms the standard Ensemble learning model in AUC-ROC, Accuracy, Precision, Recall, F1-Score, and Specificity. The enhanced model has an average accuracy of 86%, compared to 76% for the standard ensemble model across all datasets. The optimized model outperformed the conventional ensemble for the same datasets, with an average AUC of 72% compared to 61% for the standard ensemble. The optimized model was more stable than the standard model, with an STD of 5.53E−03 vs 7.24E−02 for the ensemble model. The WSO optimization process strengthens and generalizes optimizeels. The study suggests that evolutionary metaheuristic approaches can enhance EM models’ accuracy, trustworthiness, and adaptability.https://doi.org/10.1007/s44196-024-00716-0Binary white shark optimizerEnsemble learningSoftware defect classificationSoftware quality |
| spellingShingle | Jameel Saraireh Mary Agoyi Sofian Kassaymeh Adaptive Ensemble Learning Model-Based Binary White Shark Optimizer for Software Defect Classification International Journal of Computational Intelligence Systems Binary white shark optimizer Ensemble learning Software defect classification Software quality |
| title | Adaptive Ensemble Learning Model-Based Binary White Shark Optimizer for Software Defect Classification |
| title_full | Adaptive Ensemble Learning Model-Based Binary White Shark Optimizer for Software Defect Classification |
| title_fullStr | Adaptive Ensemble Learning Model-Based Binary White Shark Optimizer for Software Defect Classification |
| title_full_unstemmed | Adaptive Ensemble Learning Model-Based Binary White Shark Optimizer for Software Defect Classification |
| title_short | Adaptive Ensemble Learning Model-Based Binary White Shark Optimizer for Software Defect Classification |
| title_sort | adaptive ensemble learning model based binary white shark optimizer for software defect classification |
| topic | Binary white shark optimizer Ensemble learning Software defect classification Software quality |
| url | https://doi.org/10.1007/s44196-024-00716-0 |
| work_keys_str_mv | AT jameelsaraireh adaptiveensemblelearningmodelbasedbinarywhitesharkoptimizerforsoftwaredefectclassification AT maryagoyi adaptiveensemblelearningmodelbasedbinarywhitesharkoptimizerforsoftwaredefectclassification AT sofiankassaymeh adaptiveensemblelearningmodelbasedbinarywhitesharkoptimizerforsoftwaredefectclassification |