Improving cross-project software defect prediction method through transformation and feature selection approach
In a practical situation where the project to be predicted is new, traditional software defect prediction cannot be employed. An alternative method is cross-project defect prediction, where the historical record of one project (source) is used to predict the defect status of another project (target)...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Institute of Electrical and Electronics Engineers Inc.
2023
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/110307/ http://psasir.upm.edu.my/id/eprint/110307/1/Improving_Cross-Project_Software_Defect_Prediction_Method_Through_Transformation_and_Feature_Selection_Approach.pdf |
| _version_ | 1848865490236080128 |
|---|---|
| author | Bala, Yahaya Zakariyau Abdul Samat, Pathiah Sharif, Khaironi Yatim Manshor, Noridayu |
| author_facet | Bala, Yahaya Zakariyau Abdul Samat, Pathiah Sharif, Khaironi Yatim Manshor, Noridayu |
| author_sort | Bala, Yahaya Zakariyau |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | In a practical situation where the project to be predicted is new, traditional software defect prediction cannot be employed. An alternative method is cross-project defect prediction, where the historical record of one project (source) is used to predict the defect status of another project (target). The cross-project defect prediction method solves the limitations of the historical records in the traditional software defect prediction method. However, the performance of cross-project defect prediction is relatively low because of the distribution differences between the source and target projects. Furthermore, the software defect dataset used for cross-project defect prediction is characterized by high-dimensional features, some of which are irrelevant and contribute to low performance. To resolve these two issues, this study proposes a transformation and feature selection approach to reduce the distribution difference and high-dimensional features in cross-project defect prediction. A comparative experiment was conducted on publicly available datasets from the AEEEM. Analysis of the results obtained shows that the proposed approach in conjugation with random forest as the classification model outperformed the other four state-of-the-art cross-project defect prediction methods based on the commonly used performance evaluation metric F1score. |
| first_indexed | 2025-11-15T14:05:32Z |
| format | Article |
| id | upm-110307 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T14:05:32Z |
| publishDate | 2023 |
| publisher | Institute of Electrical and Electronics Engineers Inc. |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1103072024-09-04T03:57:25Z http://psasir.upm.edu.my/id/eprint/110307/ Improving cross-project software defect prediction method through transformation and feature selection approach Bala, Yahaya Zakariyau Abdul Samat, Pathiah Sharif, Khaironi Yatim Manshor, Noridayu In a practical situation where the project to be predicted is new, traditional software defect prediction cannot be employed. An alternative method is cross-project defect prediction, where the historical record of one project (source) is used to predict the defect status of another project (target). The cross-project defect prediction method solves the limitations of the historical records in the traditional software defect prediction method. However, the performance of cross-project defect prediction is relatively low because of the distribution differences between the source and target projects. Furthermore, the software defect dataset used for cross-project defect prediction is characterized by high-dimensional features, some of which are irrelevant and contribute to low performance. To resolve these two issues, this study proposes a transformation and feature selection approach to reduce the distribution difference and high-dimensional features in cross-project defect prediction. A comparative experiment was conducted on publicly available datasets from the AEEEM. Analysis of the results obtained shows that the proposed approach in conjugation with random forest as the classification model outperformed the other four state-of-the-art cross-project defect prediction methods based on the commonly used performance evaluation metric F1score. Institute of Electrical and Electronics Engineers Inc. 2023-01 Article PeerReviewed text en http://psasir.upm.edu.my/id/eprint/110307/1/Improving_Cross-Project_Software_Defect_Prediction_Method_Through_Transformation_and_Feature_Selection_Approach.pdf Bala, Yahaya Zakariyau and Abdul Samat, Pathiah and Sharif, Khaironi Yatim and Manshor, Noridayu (2023) Improving cross-project software defect prediction method through transformation and feature selection approach. IEEE Access, 11. pp. 2318-2326. ISSN 2169-3536 https://ieeexplore.ieee.org/document/9996402/authors#authors 10.1109/access.2022.3231456 |
| spellingShingle | Bala, Yahaya Zakariyau Abdul Samat, Pathiah Sharif, Khaironi Yatim Manshor, Noridayu Improving cross-project software defect prediction method through transformation and feature selection approach |
| title | Improving cross-project software defect prediction method through transformation and feature selection approach |
| title_full | Improving cross-project software defect prediction method through transformation and feature selection approach |
| title_fullStr | Improving cross-project software defect prediction method through transformation and feature selection approach |
| title_full_unstemmed | Improving cross-project software defect prediction method through transformation and feature selection approach |
| title_short | Improving cross-project software defect prediction method through transformation and feature selection approach |
| title_sort | improving cross-project software defect prediction method through transformation and feature selection approach |
| url | http://psasir.upm.edu.my/id/eprint/110307/ http://psasir.upm.edu.my/id/eprint/110307/ http://psasir.upm.edu.my/id/eprint/110307/ http://psasir.upm.edu.my/id/eprint/110307/1/Improving_Cross-Project_Software_Defect_Prediction_Method_Through_Transformation_and_Feature_Selection_Approach.pdf |