Defect prediction approaches have been greatly contributing to software quality assurance activities such as code review or unit testing. Just-in-time defect prediction approaches are developed to predict whether a commit is a defect-inducing commit or not. Prior research has shown that commit-level prediction is not enough in terms of effort, and a defective commit may contain both defective and non-defective files. As the defect prediction community is promoting fine-grained granularity prediction approaches, we propose our novel class-level prediction, which is finer-grained than the file-level prediction, based on the files of the commits in this research. We designed our model for Python projects and tested it with ten open-source Python projects. We performed our experiment with two settings: setting with product metrics only and setting with product metrics plus commit information. Our investigation was conducted with three different classifiers and two validation strategies. We found that our model developed by random forest classifier performs the best, and commit information contributes significantly to the product metrics in 10-fold cross-validation. We also created a commit-based file-level prediction for the Python files which do not have the classes. The file-level model also showed a similar condition as the class-level model. However, the results showed a massive deviation in time-series validation for both levels and the challenge of predicting Python classes and files in a realistic scenario.

This paper focuses on the “data collection period” for training a better Just-In-Time (JIT) defect prediction model — the early commit data vs. the recent one —, and conducts a large-scale comparative study to explore an appropriate data collection period. Since there are many possible machine learning algorithms for training defect prediction models, the selection of machine learning algorithms can become a threat to validity. Hence, this study adopts the automatic machine learning method to mitigate the selection bias in the comparative study. The empirical results using 122 open-source software projects prove the trend that the dataset composed of the recent commits would become a better training set for JIT defect prediction models.

Cross-project defect prediction (CPDP) is a hot research topic in recent years. The inconsistent data distribution between source and target projects and lack of labels for most of target instances bring a challenge for defect prediction. Researchers have developed several CPDP methods. However, the prediction performance still needs to be improved. In this paper, we propose a novel approach called Joint Domain Adaption and Pseudo-Labeling (JDAPL). The network architecture consists of a feature mapping sub-network to map source and target instances into a common subspace, followed by a classification sub-network and an auxiliary classification sub-network. The classification sub-network makes use of the label information of labeled instances to generate pseudo-labels. The auxiliary classification sub-network learns to reduce the distribution difference and improve the accuracy of pseudo-labels for unlabeled instances through loss maximization. Network training is guided by the adversarial scheme. Extensive experiments are conducted on 10 projects of the AEEEM and NASA datasets, and the results indicate that our approach achieves better performance compared with the baselines.

Various software fault prediction models have been proposed in the past twenty years. Many studies have compared and evaluated existing prediction approaches in order to identify the most effective ones. However, in most cases, such models and techniques provide varying results, and their outcomes do not result in best possible performance across different datasets. This is mainly due to the diverse nature of software development projects, and therefore, there is a risk that the selected models lead to inconsistent results across multiple datasets. In this work, we propose the use of bandit algorithms in cases where the accuracy of the models are inconsistent across multiple datasets. In the experiment discussed in this work, we used four conventional prediction models, tested on three different dataset, and then selected the best possible model dynamically by applying bandit algorithms. We then compared our results with those obtained using majority voting. As a result, Epsilon-greedy with ϵ=0.3 showed the best or second-best prediction performance compared with using only one prediction model and majority voting. Our results showed that bandit algorithms can provide promising outcomes when used in fault prediction.

Cross-project defect prediction (CPDP) is a research hot recently, which utilizes the data form existing source project to construct prediction model and predicts the defect-prone of software instances from target project. However, it is challenging in bridging the distribution difference between different projects. To minimize the data distribution differences between different projects and predict unlabeled target instances, we present a novel approach called selective pseudo-labeling based subspace learning (SPSL). SPSL learns a common subspace by using both labeled source instances and pseudo-labeled target instances. The accuracy of pseudo-labeling is promoted by iterative selective pseudo-labeling strategy. The pseudo-labeled instances from target project are iteratively updated by selecting the instances with high confidence from two pseudo-labeling technologies. Experiments are conducted on AEEEM dataset and the results show that SPSL is effective for CPDP.

To enhance the prediction accuracy of the number of faults, many studies proposed various prediction models. The model is built using a dataset collected in past projects, and the number of faults is predicted using the model and the data of the current project. Datasets sometimes have many data points where the dependent variable, i.e., the number of faults is zero. When a multiple linear regression model is made using the dataset, the model may not be built properly. To avoid the problem, the Tobit model is considered to be effective when predicting software faults. The model assumes that the range of a dependent variable is limited and the model is built based on the assumption. Similar to the Tobit model, the Poisson regression model assumes there are many data points whose value is zero on the dependent variable. Also, log-transformation is sometimes applied to enhance the accuracy of the model. Additionally, ensemble methods are effective to enhance prediction accuracy of the models. We evaluated the prediction accuracy of the methods separately, when the number of faults is zero and not zero. In the experiment, our proposed ensemble method showed the highest accuracy, and Pred25 was 21% when the number of faults was not zero, and it was 45% when the number was zero.

Software defect prediction (SDP) plays a vital role in allocating testing resources reasonably and ensuring software quality. When there are not enough labeled historical modules, considerable semi-supervised SDP methods have been proposed, and these methods utilize limited labeled modules and abundant unlabeled modules simultaneously. Nevertheless, most of them make use of traditional features rather than the powerful deep feature representations. Besides, the cost of the misclassification of the defective modules is higher than that of defect-free ones, and the number of the defective modules for training is small. Taking the above issues into account, we propose a cost-sensitive and sparse ladder network (CSLN) for SDP. We firstly introduce the semi-supervised ladder network to extract the deep feature representations. Besides, we introduce the cost-sensitive learning to set different misclassification costs for defective-prone and defect-free-prone instances to alleviate the class imbalance problem. A sparse constraint is added on the hidden nodes in ladder network when the number of hidden nodes is large, which enables the model to find robust structures of the data. Extensive experiments on the AEEEM dataset show that the CSLN outperforms several state-of-the-art semi-supervised SDP methods.

Heterogeneous defect prediction (HDP) is to detect the largest number of defective software modules in one project by using historical data collected from other projects with different metrics. However, these data can not be directly used because of different metrics set among projects. Meanwhile, software data have more non-defective instances than defective instances which may cause a significant bias towards defective instances. To completely solve these two restrictions, we propose unsupervised deep domain adaptation approach to build a HDP model. Specifically, we firstly map the data of source and target projects into a unified metric representation (UMR). Then, we design a simple neural network (SNN) model to deal with the heterogeneous and class-imbalanced problems in software defect prediction (SDP). In particular, our model introduces the Maximum Mean Discrepancy (MMD) as the distance between the source and target data to reduce the distribution mismatch, and use the cross-entropy loss function as the classification loss. Extensive experiments on 18 public projects from four datasets indicate that the proposed approach can build an effective prediction model for heterogeneous defect prediction (HDP) and outperforms the related competing approaches.

Software defect prediction (SDP) plays a significant part in allocating testing resources reasonably, reducing testing costs, and ensuring software quality. One of the most widely used algorithms of SDP models is Naive Bayes (NB) because of its simplicity, effectiveness and robustness. In NB, when a data set has continuous or numeric attributes, they are generally assumed to follow normal distributions and incorporate the probability density function of normal distribution into their conditional probabilities estimates. However, after conducting a Kolmogorov-Smirnov test, we find that the 21 main software metrics follow non-normal distribution at the 5% significance level. Therefore, this paper proposes an improved NB approach, which estimates the conditional probabilities of NB with kernel density estimation of training data sets, to help improve the prediction accuracy of NB for SDP. To evaluate the proposed method, we carry out experiments on 34 software releases obtained from 10 open source projects provided by PROMISE repository. Four well-known classification algorithms are included for comparison, namely Naive Bayes, Support Vector Machine, Logistic Regression and Random Tree. The obtained results show that this new method is more successful than the four well-known classification algorithms in the most software releases.

Association rule mining discovers relationships among variables in a data set, representing them as rules. These are expected to often have predictive abilities, that is, to be able to predict future events, but commonly used rule interestingness measures, such as support and confidence, do not directly assess their predictive power. This paper proposes a cross-validation -based metric that quantifies the predictive power of such rules for characterizing software defects. The results of evaluation this metric experimentally using four open-source data sets (Mylyn, NetBeans, Apache Ant and jEdit) show that it can improve rule prioritization performance over conventional metrics (support, confidence and odds ratio) by 72.8% for Mylyn, 15.0% for NetBeans, 10.5% for Apache Ant and 0 for jEdit in terms of SumNormPre(100) precision criterion. This suggests that the proposed metric can provide better rule prioritization performance than conventional metrics and can at least provide similar performance even in the worst case.

An transfer learning method, called Kernel Canonical Correlation Analysis plus (KCCA+), is proposed for heterogeneous Cross-company defect prediction. Combining the kernel method and transfer learning techniques, this method improves the performance of the predictor with more adaptive ability in nonlinearly separable scenarios. Experiments validate its effectiveness.

Class imbalance has drawn much attention of researchers in software defect prediction. In practice, the performance of defect prediction models may be affected by the class imbalance problem. In this paper, we present an approach to evaluating the performance stability of defect prediction models on imbalanced datasets. First, random sampling is applied to convert the original imbalanced dataset into a set of new datasets with different levels of imbalance ratio. Second, typical prediction models are selected to make predictions on these new constructed datasets, and Coefficient of Variation (C·V) is used to evaluate the performance stability of different models. Finally, an empirical study is designed to evaluate the performance stability of six prediction models, which are widely used in software defect prediction. The results show that the performance of C4.5 is unstable on imbalanced datasets, and the performance of Naive Bayes and Random Forest are more stable than other models.

An asymmetric classifier based on kernel partial least squares is proposed for software defect prediction. This method improves the prediction performance on imbalanced data sets. The experimental results validate its effectiveness.

An active learning method, called Two-stage Active learning algorithm (TAL), is developed for software defect prediction. Combining the clustering and support vector machine techniques, this method improves the performance of the predictor with less labeling effort. Experiments validate its effectiveness.

A kernel based asymmetric learning method is developed for software defect prediction. This method improves the performance of the predictor on class imbalanced data, since it is based on kernel principal component analysis. An experiment validates its effectiveness.