A deep ensemble learning method for effort-aware just-in-time defect prediction

• Autorzy: Albahli Saleh
• Języki publikacji: EN

Abstrakty

EN

Since the introduction of Just-in-Time effort aware defect prediction, many researchers are focusing on evaluating the different learning methods for defect prediction. To predict the changes that are defect-inducing, it is im-portant for learning model to consider the nature of the dataset, its imbalance properties and the correlation between different attributes. In this paper, we evaluated the importance of dataset properties, and proposed a novel methodology for learning the effort aware just-in-time defect prediction model. We form an ensemble classifier, which consider the output of three individuals classifier i.e. Random forest, XGBoost and Deep Neural Network. Our proposed methodology shows better performance with 77% accuracy on sample dataset and 81% accuracy on different dataset.

Słowa kluczowe

EN

deep neural network; unlabeled dataset; Just-In-Time defect prediction; unsupervised prediction

PL

Deep Neural Network; nieoznakowany zbiór danych; przewidywanie defektów Just-In-Time; przewidywanie bez nadzoru

• Wydawca: Polskie Towarzystwo Promocji Wiedzy ; Lublin University of Technology
• Czasopismo: Applied Computer Science
• Rocznik: 2020
• Tom: Vol. 16, no 3
• Strony: 5--15
• Opis fizyczny: Bibliogr. 11 poz., fig., tab.

Twórcy

autor

Albahli Saleh

• Qassim University, College of Computer, Department of Information Technology, Saudi Arabia, 51452, Qassim

Bibliografia

• [1] Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785–794). ACM. https://doi.org/10.1145/2939672.2939785
• [2] Hata, H., Mizuno, O., & Kikuno, T. (2012). Bug prediction based on fine-grained module histories. In Proceedings of the 34th International Conference on Software Engineering (pp. 200–210). IEEE Press.
• [3] Huang, Q., Xia, X., & Lo, D. (2019). Revisiting supervised and unsupervised models for effort-aware just-in-time defect prediction. Empirical Software Engineering, 24(5), 2823–2862. https://doi.org/10.1007/s10664-018-9661-2
• [4] Kamei, Y., Matsumoto, S., Monden, A., Matsumoto, K.I., Adams, B., & Hassan, A. E. (2010). Revisiting common bug prediction findings using effort-aware models. In 2010 IEEE International Conference on Software Maintenance (pp. 1–10). IEEE. https://doi.org/10.1109/ICSM.2010.5609530
• [5] Kamei, Y., Shihab, E., Adams, B., Hassan, A.E., Mockus, A., Sinha, A., & Ubayashi, N. (2012). A large-scale empirical study of just-in-time quality assurance. IEEE Transactions on Software Engineering, 39(6), 57–773. http://doi.org/10.1109/TSE.2012.70
• [6] Liu, C., Yang, D., Xia, X., Yan, M., & Zhang, X. (2018). Cross-Project Change-Proneness Prediction. In 2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC) (Vol. 1, pp. 64–73). IEEE.
• [7] Mockus, A., & Weiss, D.M. (2000). Predicting risk of software changes. Bell Labs Technical Journal, 5(2), 169–180.
• [8] Qiao, L., & Wang, Y. (2019). Effort-aware and just-in-time defect prediction with neural network. PloS one, 14(2), e0211359. https://doi.org/10.1371/journal.pone.0211359
• [9] Yang, Y., Zhou, Y., Liu, J., Zhao, Y., Lu, H., Xu, L., ... & Leung, H. (2016). Effort-aware just-in-time defect prediction: simple unsupervised models could be better than supervised models. In Proceedings of the 2016 24th ACM SIGSOFT International Symposium on Foundations of Software Engineering (pp. 157–168). ACM. https://doi.org/10.1145/2950290.2950353
• [10] Yu, T., Wen, W., Han, X., & Hayes, J. (2018). ConPredictor: Concurrency Defect Prediction in Real-World Applications. IEEE Transactions on Software Engineering, 45(6), 558–575. https://doi.org/10.1109/TSE.2018.2791521
• [11] Zhou, T., Sun, X., Xia, X., Li, B., & Chen, X. (2019). Improving defect prediction with deep forest. Information and Software Technology, 114, 204–216. https://doi.org/10.1016/j.infsof.2019.07.003

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2020).