An effective feature selection based cross-project defect prediction model for software quality improvement

Yogita Khatri () and Sandeep Kumar Singh ()
Additional contact information
Yogita Khatri: Jaypee Institute of Information Technology
Sandeep Kumar Singh: Jaypee Institute of Information Technology

International Journal of System Assurance Engineering and Management, 2023, vol. 14, issue 1, No 11, 154-172

Abstract: Abstract Cross-project defect prediction (CPDP) involves the use of other projects (aka source projects) for training and persuasive model building for a particular project (aka target project). However, the distribution dissimilarity between the two different project data often limits the CPDP model’s capability. Several CPDP approaches have been proposed in the literature to combat this distribution gap through instance selection by transferring the knowledge learned from the source to the target project. However, very few have explored transferring knowledge through feature selection (FS). A novel CPDP approach has been proposed consisting of two distinct FS strategies (one non-iterative and one iterative) having a trade-off between the cost and the performance respectively. The first strategy MIC_SM_FS is a non-iterative strategy that selects features that are important and have similar distribution with the corresponding target feature. The feature importance is measured using maximal information coefficient and the feature distribution similarity is calculated using 10 statistical measures. On the other hand, the second strategy BPSO_FS is an iterative strategy that works on optimizing the performance, utilizing the powerful binary particle swarm optimization algorithm for selecting the representative features for CPDP. Both of the proposed strategies have been tested on 26 cross-project experiments based on 8 software projects. From the two proposed strategies, a CPDP model built utilizing BPSO_FS showed better results. Further, to assess its performance, comparison is done with two baseline approaches viz. ALL and ManualDown, within-project defect prediction, and a state-of-the-art CPDP technique TCA+. Statistical results showed the potential of the proposed CPDP approach over the compared approaches.

Keywords: Cross-project defect prediction; Maximal information coefficient; Binary particle swarm optimization algorithm; Feature selection (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s13198-022-01831-x Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:ijsaem:v:14:y:2023:i:1:d:10.1007_s13198-022-01831-x

Ordering information: This journal article can be ordered from
http://www.springer.com/engineering/journal/13198

DOI: 10.1007/s13198-022-01831-x

Access Statistics for this article

International Journal of System Assurance Engineering and Management is currently edited by P.K. Kapur, A.K. Verma and U. Kumar

More articles in International Journal of System Assurance Engineering and Management from Springer, The Society for Reliability, Engineering Quality and Operations Management (SREQOM),India, and Division of Operation and Maintenance, Lulea University of Technology, Sweden
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().