Online Defect Prediction for Imbalanced Data

Abstract

Abstract-Many defect prediction techniques are proposed to improve software reliability. Change classification predicts defects at the change level, where a change is the modifications to one file in a commit. In this paper, we conduct the first study of applying change classification in practice.We identify two issues in the prediction process, both of which contribute to the low prediction performance. First, the data are imbalanced-there are much fewer buggy changes than clean changes. Second, the commonly used cross-validation approach is inappropriate for evaluating the performance of change classification. To address these challenges, we apply and adapt online change classification, resampling, and updatable classification techniques to improve the classification performance.We perform the improved change classification techniques on one proprietary and six open source projects. Our results show that these techniques improve the precision of change classification by 12.2-89.5% or 6.4-34.8 percentage points (pp.) on the seven projects. In addition, we integrate change classification in the development process of the proprietary project. We have learned the following lessons: 1) new solutions are needed to convince developers to use and believe prediction results, and prediction results need to be actionable, 2) new and improved classification algorithms are needed to explain the prediction results, and insensible and unactionable explanations need to be filtered or refined, and 3) new techniques are needed to improve the relatively low precision.

Towards an ensemble based system for predicting the number of software faults

“…These studies showed that oversampling approach helps in achieving better prediction performance when dataset is imbalanced. In addition, most of the reported works (e.g., Li & Wang, 2014;Pelayo & Dick, 2007;Shatnawi, 2012;Tan, Tan, Dara, & Mayeux, 2015 ) on software fault prediction have used oversampling approach for generating synthetic values. Due to these reasons, we have used an oversampling approach in this study.…”

Adressing problems with external validity of repository mining studies through a smart data platform

“…Using the functionality provided by Spark, creating the data splits as required for the approach by Tan et al [71] as well as the training of the defect prediction model was straightforward. After fetching the data from the MongoDB, a Map job was used to prepare the data.…”

“…We selected a change-based defect prediction model based on a recent publication by Tan et al [71]. The approach by Tan et al suggests to use the first part of a project as training data, then leave a gap and predict the remainder of the project using a prediction model trained on the first part of the data.…”

Heterogeneous fault prediction with cost‐sensitive domain adaptation

“…For addressing the class imbalance problem in fault prediction, numerous methods have been developed at data and algorithm levels. Data‐level methods include a variety of resampling techniques, such as random undersampling, random oversampling, and SMOTE (Synthetic Minority Over‐sampling TEchnique) .…”