- Title
- Cross-project and within-project semi-supervised software defect prediction problems study using a unified solution
- Creator
- Wu, Fei; Jing, Xiao-Yuan; Dong, Xiwei; Cao, Jicheng; Xu, Mingwei; Zhang, Hongyu; Ying, Shi; Xu, Baowen
- Relation
- 39th International Conference on Software Engineering Companion (ICSE-C 2017). Proceedings of the 39th International Conference on Software Engineering Companion (Buenos Aires, Argentina 20-28 May, 2017) p. 195-197
- Publisher Link
- http://dx.doi.org/10.1109/ICSE-C.2017.72
- Publisher
- Institute of Electrical and Electronics Engineers (IEEE)
- Resource Type
- conference paper
- Date
- 2017
- Description
- When there exists not enough historical defect data for building accurate prediction model, semi-supervised defect prediction (SSDP) and cross-project defect prediction (CPDP) are two feasible solutions. Existing CPDP methods assume that the available source data is well labeled. However, due to expensive human efforts for labeling a large amount of defect data, usually, we can only make use of the suitable unlabeled source data to help build the prediction model. We call CPDP in this scenario as cross-project semi-supervised defect prediction (CSDP). As to within-project semi-supervised defect prediction (WSDP), although some WSDP methods have been developed in recent years, there still exists much room for improvement. In this paper, we aim to provide an effective solution for both CSDP and WSDP problems. We introduce the semi-supervised dictionary learning technique, an effective machine learning technique, into defect prediction and propose a semi-supervised structured dictionary learning (SSDL) approach for CSDP and WSDP. SSDL can make full use of the useful information in limited labeled defect data and a large amount of unlabeled data. Experiments on two public datasets indicate that SSDL can obtain better prediction performance than related SSDP methods in the CSDP scenario.
- Subject
- cross-project semi-supervised defect prediction; within-project semi-supervised defect prediction; semi-supervised structured dictionary learning
- Identifier
- http://hdl.handle.net/1959.13/1385340
- Identifier
- uon:32208
- Identifier
- ISBN:9781538615898
- Language
- eng
- Reviewed
- Hits: 3802
- Visitors: 3799
- Downloads: 0
| Thumbnail | File | Description | Size | Format |
|---|