A novel modified undersampling (MUS) technique for software defect prediction

KCO: Balancing class distribution in just-in-time software defect prediction using kernel crossover oversampling

The performance of the defect prediction model by using balanced and imbalanced datasets makes a big impact on the discovery of future defects. Current resampling techniques only address the imbalanced datasets without taking into consideration redundancy and noise inherent to the imbalanced datasets. To address the imbalance issue, we propose Kernel Crossover Oversampling (KCO), an oversampling technique based on kernel analysis and crossover interpolation. Specifically, the proposed technique aims to generate balanced datasets by increasing data diversity in order to reduce redundancy and noise. KCO first represents multidimensional features into two-dimensional features by employing Kernel Principal Component Analysis (KPCA). KCO then divides the plotted data distribution by deploying spectral clustering to select the best region for interpolation. Lastly, KCO generates the new defect data by interpolating different data templates within the selected data clusters. According to the prediction evaluation conducted, KCO consistently produced F-scores ranging from 21% to 63% across six datasets, on average. According to the experimental results presented in this study, KCO provides more effective prediction performance than other baseline techniques. The experimental results show that KCO within project and cross project predictions especially consistently achieve higher performance of F-score results.

Predicting defects in imbalanced data using resampling methods: an empirical investigation

The development of correct and effective software defect prediction (SDP) models is one of the utmost needs of the software industry. Statistics of many defect-related open-source data sets depict the class imbalance problem in object-oriented projects. Models trained on imbalanced data leads to inaccurate future predictions owing to biased learning and ineffective defect prediction. In addition to this large number of software metrics degrades the model performance. This study aims at (1) identification of useful metrics in the software using correlation feature selection, (2) extensive comparative analysis of 10 resampling methods to generate effective machine learning models for imbalanced data, (3) inclusion of stable performance evaluators-AUC, GMean, and Balance and (4) integration of statistical validation of results. The impact of 10 resampling methods is analyzed on selected features of 12 object-oriented Apache datasets using 15 machine learning techniques. The performances of developed models are analyzed using AUC, GMean, Balance, and sensitivity. Statistical results advocate the use of resampling methods to improve SDP. Random oversampling portrays the best predictive capability of developed defect prediction models. The study provides a guideline for identifying metrics that are influential for SDP. The performances of oversampling methods are superior to undersampling methods.

RDPVR: Random Data Partitioning with Voting Rule for Machine Learning from Class-Imbalanced Datasets

Since most classifiers are biased toward the dominant class, class imbalance is a challenging problem in machine learning. The most popular approaches to solving this problem include oversampling minority examples and undersampling majority examples. Oversampling may increase the probability of overfitting, whereas undersampling eliminates examples that may be crucial to the learning process. We present a linear time resampling method based on random data partitioning and a majority voting rule to address both concerns, where an imbalanced dataset is partitioned into a number of small subdatasets, each of which must be class balanced. After that, a specific classifier is trained for each subdataset, and the final classification result is established by applying the majority voting rule to the results of all of the trained models. We compared the performance of the proposed method to some of the most well-known oversampling and undersampling methods, employing a range of classifiers, on 33 benchmark machine learning class-imbalanced datasets. The classification results produced by the classifiers employed on the generated data by the proposed method were comparable to most of the resampling methods tested, with the exception of SMOTEFUNA, which is an oversampling method that increases the probability of overfitting. The proposed method produced results that were comparable to the Easy Ensemble (EE) undersampling method. As a result, for solving the challenge of machine learning from class-imbalanced datasets, we advocate using either EE or our method.