A compositional model for effort‐aware Just‐In‐Time defect prediction on android apps

Abstract

Android apps have played important roles in daily life and work. To meet the new requirements from users, the apps encounter frequent updates, which involves a large quantity of code commits. Previous studies proposed to apply Just‐in‐Time (JIT) defect prediction for apps to timely identify whether the new code commits can introduce defects into apps, aiming to assure their quality. In general, high‐quality features are benefits for improving the classification performance. In addition, the number of defective commit instances is much fewer than that of clean ones, that is the defect data is class imbalanced. In this study, a novel compositional model, called KPIDL, is proposed to conduct the JIT defect prediction task for Android apps. More specifically, KPIDL first exploits a feature learning technique to preprocess original data for obtaining better feature representation, and then introduces a state‐of‐the‐art cost‐sensitive cross‐entropy loss function into the deep neural network to alleviate the class imbalance issue by considering the prior probability of the two types of classes. The experiments were conducted on a benchmark defect data consisting of 15 Android apps. The experimental results show that the proposed KPIDL model performs significantly better than 25 comparative methods in terms of two effort‐aware performance indicators in most cases.

On the use of deep learning in software defect prediction

“…Four studies, i.e., Sheng et al (2020), , Zhu et al (2020), and Wang et al (2021), used "PofB20" (Jiang et al, 2013) to measure the percentage of defects that a developer can identify by inspecting the top 20% lines of code. Four studies, i.e., Qiao & Wang (2019), Xu et al (2019), Xu et al (2021b), andZhao et al (2021b), Effort-Aware recall (EARecall), which is defined as the percent of reviewed defective commit instances to the whole defective commit instances. Three studies, i.e., Xu et al (2019), Xu et al (2021b), andZhao et al (2021b), Effort-Aware F-measure (EAF-measure), which is defined as the weighted harmonic average between EARecall and EAPrecision.…”

“…Four studies, i.e., Qiao & Wang (2019), Xu et al (2019), Xu et al (2021b), andZhao et al (2021b), Effort-Aware recall (EARecall), which is defined as the percent of reviewed defective commit instances to the whole defective commit instances. Three studies, i.e., Xu et al (2019), Xu et al (2021b), andZhao et al (2021b), Effort-Aware F-measure (EAF-measure), which is defined as the weighted harmonic average between EARecall and EAPrecision. Xu et al (2019) used EAPrecision in addition to EARecall and EAF-measure.…”

“…Feature redundancy: Highly correlating features may reduce prediction performance (Tran et al, 2019;Turabieh et al, 2019;Wei et al, 2019;Zhao et al, 2021b;Zhu et al, 2021b). Solutions include using deep learning architectures and ensemble techniques which can cope with this and using meta-heuristic approaches for feature selection.…”

“…Feature redundancy, i.e., highly correlated features, may harm prediction performance (Tran et al, 2019;Turabieh et al, 2019;Wei et al, 2019;Zhao et al, 2021b;…”

Deep Learning-Based Defect Prediction for Mobile Applications

“…Zhao et al (2022) proposed a model for imbalanced datasets of 15 Android applications. They used feature learning with loss function into deep learning for imbalance issue and their model performance was better when compared to 25 defect prediction models [19]. Dong et al (2017) proposed a study on the defect prediction of Android apk files.…”