Software Defect Prediction Research Articles

Enhancing software defect prediction models using metaheuristics with a learning to rank approach

Enhancing software defect prediction models using metaheuristics with a learning to rank approach

An ensemble model for addressing class imbalance and class overlap in software defect prediction

An ensemble model for addressing class imbalance and class overlap in software defect prediction

Visual software defect prediction method based on improved recurrent criss-cross residual network

Purpose This study aims to solve the problems of large training sample size, low data sample quality, low efficiency of the currently used classical model, high computational complexity of the existing concern mechanism, and high graphics processing unit (GPU) occupancy in the current visualization software defect prediction, proposing a method for software defect prediction termed recurrent criss-cross attention for weighted activation functions of recurrent SE-ResNet (RCCA-WRSR). First, following code visualization, the activation functions of the SE-ResNet model are replaced with a weighted combination of Relu and Elu to enhance model convergence. Additionally, an SE module is added before it to filter feature information, eliminating low-weight features to generate an improved residual network model, WRSR. To focus more on contextual information and establish connections between a pixel and those not in the same cross-path, the visualized red as integer, green as integer, blue as integer images are inputted into a model incorporating a fused RCCA module for defect prediction. Design/methodology/approach Software defect prediction based on code visualization is a new software defect prediction technology, which mainly realizes the defect prediction of code by visualizing code as image, and then applying attention mechanism to extract the features of image. However, the challenges of current visualization software defect prediction mainly include the large training sample size and low sample quality of the data, and the classical models used today are not efficient, and the existing attention mechanisms have high computational complexity and high GPU occupancy. Findings Experimental evaluation using ten open-source Java data sets from PROMISE and five existing methods demonstrates that the proposed approach achieves an F-measure value of 0.637 in predicting 16 cross-version projects, representing a 6.1% improvement. Originality/value RCCA-WRSR is a new visual software defect prediction based on recurrent criss-cross attention and improved residual network. This method effectively enhances the performance of software defect prediction.

The effect of data complexity on classifier performance

The research area of Software Defect Prediction (SDP) is both extensive and popular, and is often treated as a classification problem. Improvements in classification, pre-processing and tuning techniques, (together with many factors which can influence model performance) have encouraged this trend. However, no matter the effort in these areas, it seems that there is a ceiling in the performance of the classification models used in SDP. In this paper, the issue of classifier performance is analysed from the perspective of data complexity. Specifically, data complexity metrics are calculated using the Unified Bug Dataset, a collection of well-known SDP datasets, and then checked for correlation with the defect prediction performance of machine learning classifiers (in particular, the classifiers C5.0, Naive Bayes, Artificial Neural Networks, Random Forests, and Support Vector Machines). In this work, different domains of competence and incompetence are identified for the classifiers. Similarities and differences between the classifiers and the performance metrics are found and the Unified Bug Dataset is analysed from the perspective of data complexity. We found that certain classifiers work best in certain situations and that all data complexity metrics can be problematic, although certain classifiers did excel in some situations.

Refining software defect prediction through attentive neural models for code understanding

Identifying defects through manual software testing is a resource-intensive task in software development. To alleviate this, software defect prediction identifies code segments likely to contain faults using data-driven methods. Traditional techniques rely on static code metrics, which often fail to reflect the deeper syntactic and semantic features of the code. This paper introduces a novel framework that utilizes transformer-based networks with attention mechanisms to predict software defects. The framework encodes input vectors to develop meaningful representations of software modules. A bidirectional transformer encoder is employed to model programming languages, followed by fine-tuning with labeled data to detect defects. The performance of the framework is assessed through experiments across various software projects and compared against baseline techniques. Additionally, statistical hypothesis testing and an ablation study are performed to assess the impact of different parameter choices. The empirical findings indicate that the proposed approach can increase classification accuracy by an average of 15.93% and improve the F1 score by up to 44.26% compared to traditional methods.

A novel deep neural network structure for software fault prediction

A novel deep neural network structure for software fault prediction

A Software Defect Prediction Method That Simultaneously Addresses Class Overlap and Noise Issues after Oversampling

A Software Defect Prediction Method That Simultaneously Addresses Class Overlap and Noise Issues after Oversampling

Online cross-project approach with project-level similarity for just-in-time software defect prediction

Online cross-project approach with project-level similarity for just-in-time software defect prediction

An efficient instance selection algorithm for fast training of support vector machine for cross-project software defect prediction pairs

SVM is limited in its use for cross-project software defect prediction because of its very slow training process. So, this research article proposes a new instance selection (IS) algorithm called boundary detection among classes (BDAC) to reduce the training dataset size for faster training of SVM without degrading the prediction performance. The proposed algorithm is evaluated against six existing IS algorithms based on accuracy, running time, data reduction rate, etc. using 23 general datasets, 18 software defect prediction datasets, and two shape-based datasets, and results prove that BDAC is better than the selected algorithm based on collective comparison.

A Framework for Developing Correct Software Programs through Software Defect Prediction and Elimination using Machine Learning Models

A Framework for Developing Correct Software Programs through Software Defect Prediction and Elimination using Machine Learning Models

Analysis of ERP Based Framework Expert Opinion in Early Software Defect Prediction

Abstract: The software industry necessitates early prediction of software defects for effective quality assessment and resource allocation. During the initial stages of the software development life cycle (SDLC), failure data is often unavailable. Consequently, the insights of domain experts can be crucial in estimating potential software defects during these early phases. This paper introduces a model designed to forecast software defects prior to the testing phase, emphasizing the structure of the software development process. The model is developed using metrics derived from early artifacts of the SDLC. The development and experimental aspects of the model are presented through the application of a Bayesian belief network (BBN). The qualitative aspects of software metrics, along with expert opinions, form the core of this methodology. To demonstrate the practicality and effectiveness of the proposed approach, ten datasets from real software projects have been utilized. The analysis and validation of predicted software defects, based on varying levels of uncertainty from domain experts, are compared against actual defect occurrences.

Enhancing Software Defect Prediction with Ensemble Models based on Defect Relations Rule Learning

Enhancing Software Defect Prediction with Ensemble Models based on Defect Relations Rule Learning

Software Defect Prediction Based on Machine Learning and Deep Learning Techniques: An Empirical Approach

Software Defect Prediction Based on Machine Learning and Deep Learning Techniques: An Empirical Approach

An Empirical Study of the Impact of Class Overlap on the Performance and Interpretability of Cross-Version Defect Prediction

The class overlap problem refers to instances from different categories heavily overlapping in the feature space. This issue is one of the challenges in improving the performance of software defect prediction (SDP). Currently, the studies on the impact of class overlap on SDP mainly focused on within-project defect prediction and cross-project defect prediction. Moreover, the existing class overlap instances cleaning methods are not suitable for cross-version defect prediction. In this paper, we propose a class overlap instances cleaning method based on the Ratio of K-nearest neighbors with the Same Label (RKSL). This method removes instances with the abnormal neighbor ratio in the training set. Based on the RKSL method, we investigate the impact of class overlap on the performance and interpretability of the cross-version defect prediction model. The experiment results show that class overlap can affect the performance of cross-version defect prediction models significantly. The RKSL method can handle the class overlap problem in defect datasets, but it may impact the interpretability of models. Through the analysis of feature changes, we consider that class overlap instances cleaning can assist models in identifying more important features.

A hybrid‐ensemble model for software defect prediction for balanced and imbalanced datasets using AI‐based techniques with feature preservation: SMERKP‐XGB

AbstractMaintaining software quality is a significant challenge as the complexity of software is increasing with the rise of the software industry. Software defects are a primary concern in complex modules, and predicting them in the early stages of the software development life cycle (SDLC) is difficult. Previous techniques to address this issue have not been very promising. We have proposed “A hybrid ensemble model for software defect prediction using AI‐based techniques with feature preservation” to overcome this problem. We have used the National Aeronautics and Space Administration (NASA) dataset from the PROMISE repository for testing and validation. By applying exploratory data analysis (EDA), feature engineering, scaling, and standardization, we found that the dataset is imbalanced, which can negatively affect the model's performance. To address this, we have used the Synthetic Minority Oversampling (SMOTE) technique and the edited nearest neighbor (ENN) (SMOTE‐ENN). We have also used recursive feature elimination cross‐validation (RFE‐CV) with a pipeline to prevent data leaking in CV and kernel‐based principal component analysis (K‐PCA) to minimize dimensionality and selectively relevant features. The reduced dimensional data is then given to the eXtreme Gradient Boosting (XGBoost) for classification, resulting in the hybrid‐ensemble (SMERKP‐XGB) model. The proposed SMERKP‐XGB model is better than previously developed models in terms of accuracy (CM1: 97.53%, PC1: 92.05%, and PC2: 97.45%, KC1:95.65%), and area under the receiver operating characteristic curve values (CM1:96.30%, PC1:98.30%, and PC2:99.30%: KC1: 93.54) and other evaluation criteria mentioned in the literature.

Optimization of Backward Elimination for Software Defect Prediction with Correlation Coefficient Filter Method

Detecting software defects is a crucial step for software development not only to reduce cost and save time, but also to mitigate more costly losses. Backward Elimination is one method for detecting software defects. Notably Backward Elimination may remove features that may later become significant to the outcome affecting the performance of Backward Elimination. The aim of this study is to improve Backward Elimination performance. In this study, several features were selected based on their correlation coefficient, with the selected feature applied to improve Backward Elimination final model performance. The final model was validated using cross validation with Naïve Bayes as the classification method on the NASA MDP dataset to determine the accuracy and Area Under the Curve (AUC) of the final model. Using top 10 correlation feature and Backward Elimination achieve an average result of 86.6% accuracy and 0.797 AUC, while using top 20 correlation feature and Backward Elimination achieved an average result of 84% accuracy and 0.812 AUC. Compare to using Backward Elimination and Naïve Bayes respectively the improvement using top 10 correlation feature as follows: AUC:1.52%, 13.53% and Accuracy: 13%, 12.4% while the improvement using top 20 correlation feature as follows: AUC:3.43%, 15.66% and Accuracy: 10.4%, 9.8%. Results showed that selecting the top 10 and top 20 feature based on its correlation before using Backward Elimination have better result than only using Backward Elimination. This result shows that combining Backward Elimination with correlation coefficient feature selection does improve Backward Elimination’s final model and yielding good results for detecting software defects.

An enhanced conventional neural network schema for structural class-based software fault prediction

ABSTRACT Malicious software detection is the most prominent process required by various industries to avoid server failure. It is required to detect malicious software accurately to avoid time and cost wastage. Various research works have been introduced earlier for the detection of malicious software. In the existing work Support Vector Machine (SVM) is introduced for malicious software detection. However, existing works cannot perform well where there are error modules in the software. It is addressed in this suggested study by developing Coupling and Cohesion Metrics-based Fault Detection (CCMFD). In this research work, structural measures are mainly examined which come under the cohesion measures and comprise deficient cohesion in approaches (LCOM), and Conceptual Coupling between Object Classes (CCBO). Failure situations and measures relating to information flow are used in other techniques. A high-quality service has a low coupling and a high cohesiveness. These extracted features will be given as input to the enhanced Conventional Neural Network (CNN) for software mistake forecasting. A complete study analysis is done in a Java simulator, indicating that the suggested approach tends to have superior fault prediction outcomes than the current method.

A holistic approach to software fault prediction with dynamic classification

Software Fault Prediction is a critical domain in machine learning aimed at pre-emptively identifying and mitigating software faults. This study addresses challenges related to imbalanced datasets and feature selection, significantly enhancing the effectiveness of fault prediction models. We mitigate class imbalance in the Unified Dataset using the Random-Over Sampling technique, resulting in superior accuracy for minority-class predictions. Additionally, we employ the innovative Ant-Colony Optimization algorithm (ACO) for feature selection, extracting pertinent features to amplify model performance. Recognizing the limitations of individual machine learning models, we introduce the Dynamic Classifier, a ground-breaking ensemble that combines predictions from multiple algorithms, elevating fault prediction precision. Model parameters are fine-tuned using the Grid-Search Method, achieving an accuracy of 94.129% and superior overall performance compared to random forest, decision tree and other standard machine learning algorithms. The core contribution of this study lies in the comparative analysis, pitting our Dynamic Classifier against Standard Algorithms using diverse performance metrics. The results unequivocally establish the Dynamic Classifier as a frontrunner, highlighting its prowess in fault prediction. In conclusion, this research introduces a comprehensive and innovative approach to software fault prediction. It pioneers the resolution of class imbalance, employs cutting-edge feature selection, and introduces dynamic ensemble classifiers. The proposed methodology, showcasing a significant advancement in performance over existing methods, illuminates the path toward developing more accurate and efficient fault prediction models.

A three-step combination strategy for addressing outliers and class imbalance in software defect prediction

<p>Software defect prediction often involves datasets with imbalanced distributions where one or more classes are underrepresented, referred to as the minority class, while other classes are overrepresented, known as the majority class. This imbalance can hinder accurate predictions of the minority class, leading to misclassification. While the synthetic minority oversampling technique (SMOTE) is a widely used approach to address imbalanced learning data, it can inadvertently generate synthetic minority samples that resemble the majority class and are considered outliers. This study aims to enhance SMOTE by integrating it with an efficient algorithm designed to identify outliers among synthetic minority samples. The resulting method, called reduced outliers (RO)-SMOTE, is evaluated using an imbalanced dataset, and its performance is compared to that of SMOTE. RO-SMOTE first performs oversampling on the training data using SMOTE to balance the dataset. Next, it applies the mining outlier algorithm to detect and eliminate outliers. Finally, RO-SMOTE applies SMOTE again to rebalance the dataset before introducing it to the underlying classifier. The experimental results demonstrate that RO-SMOTE achieves higher accuracy, precision, recall, F1-score, and area under curve (AUC) values compared to SMOTE.</p>

Improving effort-aware just-in-time defect prediction with weighted code churn and multi-objective slime mold algorithm

Effort-aware just-in-time software defect prediction (JIT-SDP) aims to effectively utilize the limited resources of software quality assurance (SQA) to detect more software defects. This improves the efficiency of SQA work and the quality of the software. However, there is disagreement regarding the representation of the key feature variable, SQA effort, in the field of effort-aware JIT-SDP. Additionally, the most recent metaheuristic optimization algorithms (MOAs) have not yet been effectively integrated with multi-objective effort-aware JIT-SDP tasks. These deficiencies, in both feature representation and model optimization (MO), result in a significant disparity between the performance of effort-aware JIT-SDP techniques and the expectations of the industry. In this study, we present a novel method called weighted code churn (CC) and improved multi-objective slime mold algorithm (SMA) (WCMS) for effort-aware JIT-SDP. It comprises two stages: feature improvement (FI) and MO. In the FI phase, we normalize the two feature variables: number of modified files (NF) and distribution of modified code across each file (Entropy). We then use an exponential function to quantify the level of difficulty of the change. The equation is as follows: DD = NFEntropy, where DD is an acronym for the degree of difficulty, NF denotes the base number, and Entropy denotes the index. We define change effort as the product of the difficulty degree in implementing the change and CC, with weighted CC representing the change effort. During the MO stage, we improve the SMA by incorporating multi-objective handling capabilities and devising mechanisms for multi-objective synchronization and conflict resolution. We develop a multi-objective optimization algorithm for hyperparameter optimization (HPO) of the JIT-SDP model in WCMS. To evaluate the performance of our method, we conducted experiments using six well-known open-source projects and employed two effort-aware performance evaluation metrics. We evaluated our method based on three scenarios: cross-validation, time-wise cross-validation, and across-project prediction. The experimental results indicate that the proposed method outperforms the benchmark method. Furthermore, the proposed method demonstrates greater scalability and generalization capabilities.