Page 188 - 《软件学报》2025年第10期

李志强 等: SZZ 误标变更对移动 APP 即时缺陷预测性能和解释的影响 4585

一方面代码提交数量较多, 以确保实验中有足够的实例.

6 总结与展望

本文选取 GitHub 库中 17 个大型开源移动 APP 项目, 抽取了 Kamei 等人 [9] 提出的 14 个变更度量元, 并使用
B-SZZ、AG-SZZ、MA-SZZ 和 RA-SZZ 算法进行数据标注, 构造了 17 个移动 APP 数据集. 为探究不同 SZZ 算法
错误标注的变更对移动 APP 即时缺陷预测性能与解释的影响, 本文基于 4 种 SZZ 算法标注的数据集利用随机森
林、朴素贝叶斯和逻辑回归分类器分别建立即时缺陷预测模型, 采用 AUC、MCC、G-mean、F-measure@20%
及 IFA 这 5 个指标进行评估, 并使用 SKESD 和 SHAP 算法对结果进行排序比较与解释分析. 在模型性能方面, B-
SZZ、AG-SZZ、MA-SZZ 算法会在不同程度上导致即时缺陷预测模型性能下降. 在模型解释方面, B-SZZ、AG-
SZZ、MA-SZZ 算法会影响模型预测过程中最重要的前 3 名度量元.
在未来的即时缺陷预测研究中, 本文推荐使用 RA-SZZ 算法对数据进行标注以构建缺陷预测数据集. 此外, 本
文的预测粒度为变更级, 相比于代码行, 粒度仍然较大, 因此后续工作拟研究基于代码行级的移动 APP 即时缺陷
预测, 这将大幅度缩小开发人员需要人工审查的代码范围, 有助于进一步提升软件开发效率.

References:
[1] Chen X, Gu Q, Liu WS, Liu SL, Ni C. Survey of static software defect prediction. Ruan Jian Xue Bao/Journal of Software, 2016, 27(1):
1–25 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/4923.htm [doi: 10.13328/j.cnki.jos.004923]
[2] Gong LN, Jiang SJ, Jiang L. Research progress of software defect prediction. Ruan Jian Xue Bao/Journal of Software, 2019, 30(10):
3090–3114 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/5790.htm [doi: 10.13328/j.cnki.jos.005790]
[3] Cai L, Fan YR, Yan M, Xia X. Just-in-time software defect prediction: Literature review. Ruan Jian Xue Bao/Journal of Software, 2019,
30(5): 1288–1307 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/5713.htm [doi: 10.13328/j.cnki.jos.05713]
[4] Li ZQ, Jing XY, Zhu XK. Progress on approaches to software defect prediction. IET Software, 2018, 12(3): 161–175. [doi: 10.1049/iet-
sen.2017.0148]
[5] Fan YR, Xia X, Da Costa DA, Lo D, Hassan AE, Li SP. The impact of mislabeled changes by SZZ on just-in-time defect prediction.
IEEE Trans. on Software Engineering, 2021, 47(8): 1559–1586. [doi: 10.1109/TSE.2019.2929761]
[6] Kamei Y, Fukushima T, McIntosh S, Yamashita K, Ubayashi N, Hassan AE. Studying just-in-time defect prediction using cross-project
models. Empirical Software Engineering, 2016, 21(5): 2072–2106. [doi: 10.1007/s10664-015-9400-x]
[7] Ge J, Yu HQ, Fan GS, Tang JH, Huang ZJ. Just-in-time defect prediction for intelligent computing frameworks. Ruan Jian Xue
Bao/Journal of Software, 2023, 34(9): 3966–3980 (in Chinese with English abstract). http://www.jos.org.cn/1000-9825/6874.htm [doi: 10.
13328/j.cnki.jos.006874]
[8] Jiang T, Tan L, Kim S. Personalized defect prediction. In: Proc. of the 28th IEEE/ACM Int’l Conf. on Automated Software Engineering
(ASE). Silicon Valley: IEEE, 2013. 279–289. [doi: 10.1109/ASE.2013.6693087]
[9] Kamei Y, Shihab E, Adams B, Hassan AE, Mockus A, Sinha A, Ubayashi N. A large-scale empirical study of just-in-time quality
assurance. IEEE Trans. on Software Engineering, 2013, 39(6): 757–773. [doi: 10.1109/TSE.2012.70]
[10] Li ZQ, Zhang HY, Jing XY, Xie JY, Guo M, Ren J. DSSDPP: Data selection and sampling based domain programming predictor for
cross-project defect prediction. IEEE Trans. on Software Engineering, 2023, 49(4): 1941–1963. [doi: 10.1109/TSE.2022.3204589]
[11] Xia X, Lo D, Pan SJ, Nagappan N, Wang XY. HYDRA: Massively compositional model for cross-project defect prediction. IEEE Trans.
on Software Engineering, 2016, 42(10): 977–998. [doi: 10.1109/TSE.2016.2543218]
[12] Śliwerski J, Zimmermann T, Zeller A. When do changes induce fixes? ACM SIGSOFT Software Engineering Notes, 2005, 30(4): 1–5.
[doi: 10.1145/1082983.1083147]
[13] Kim S, Zimmermann T, Jr Pan K, Whitehead E. Automatic identification of bug-introducing changes. In: Proc. of the 21st IEEE/ACM Int’l
Conf. on Automated Software Engineering (ASE 2006). Tokyo: IEEE, 2006. 81–90. [doi: 10.1109/ASE.2006.23]
[14] Da Costa DA, McIntosh S, Shang WY, Kulesza U, Coelho R, Hassan AE. A framework for evaluating the results of the SZZ approach for
identifying bug-introducing changes. IEEE Trans. on Software Engineering, 2017, 43(7): 641–657. [doi: 10.1109/TSE.2016.2616306]
[15] Neto EC, Da Costa DA, Kulesza U. The impact of refactoring changes on the SZZ algorithm: An empirical study. In: Proc. of the 25th
IEEE Int’l Conf. on Software Analysis, Evolution and Reengineering (SANER). Campobasso: IEEE, 2018. 380–390. [doi: 10.1109/
SANER.2018.8330225]
[16] Tantithamthavorn C, Hassan AE, Matsumoto K. The impact of class rebalancing techniques on the performance and interpretation of