Software Defect Prediction Using Dictionary Learning

Abstract

Abstract-With the popularization of software version control system and defect tracking tools, large amounts of software development data is recorded. How to effectively use these data to improve the quality of software development, has become a hot topic in recent years. Software defect prediction technology can take full advantage of the historical data to build predictive models and automatically detect defective modules for efficient software test to improve the quality of a software system. But the class-imbalanced data makes the prediction model classifying a modules as a defective-free one easily, while the misclassification of defective modules generally incurs much higher cost risk than the misclassification of defective-free ones. To resolve this problem, we propose a cost-sensitive software defect prediction method using dictionary learning. It iteratively optimizes the classifier parameters and the dictionary atoms, to ensure that the extracted features (sparse representation) are optimal for the trained classifier; Moreover, we take the different misclassification costs into account, increasing the punishment on misclassification defective modules in the procedure of dictionary learning, making classification inclining to classify a module as a defective one. Experimental results on the 10 classimbalanced data sets of NASA show that our method is more effective than other methods.

Software Defect Prediction System Based on Decision Tree Algorithm

“…As a result, software defects have become a significant challenge in software development, as they can lead to significant losses, including financial losses, damage to the reputation of the company, and even loss of life in extreme cases [2]. Software defects are expensive to fix, and they can cause project delays, leading to increased costs and lost productivity [3,4]. Therefore, software defect prediction has become an essential aspect of software engineering, as it helps to identify potential defects in advance before they cause any significant issues.…”

Software Defect Prediction Using Hybrid Distribution Base Balance Instance Selection and Radial Basis Function Classifier

“…While solving the software defect prediction problem, incorporation of both labeled and unlabeled data in the machine learning process may lead to best possible classification results. To this end, many researchers have used Graph based learning with application of sparse theory on the dataset for pairwise relationship [8]; collaborative representation by the authors [9]; metrics-based [10]; class imbalance [11]; Dictionary learning [12], traditional methods like: Support Vector Machine (SVM) [13] , Naive Bayesian (NB) [14], Neural Network [15] and the list goes on. It is observed that performance of the traditional methods severely limited with respect to lack of common feature representation and selection of a good feature selection algorithm in order to deal with sparse nature of the software prediction dataset.…”