Multiple-classifiers in software quality engineering: Combining predictors to improve software fault prediction ability

Yücalar, Fatih; Özçift, Akın; Borandağ, Emin; Kılınç, Deniz

doi:10.1016/j.jestch.2019.10.005

Cited by 38 publications

(28 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results obtained after classification confirmed the improvement in performance by combining sampling techniques with an ensemble classifier. In [68], ten ensemble classifiers were compared to baseline classifiers. The evaluated ensemble learning algorithms were adaBoostM1, LogicBoost, Multiboost AB, Bagging, RF, Dagging, Rotation forest (ROF), stacking, multi scheme, and voting.…”

Section: Rq1 : Which Ensemble Learning Techniques Are Applied For Software Defect Prediction?mentioning

confidence: 99%

“…[5] performed analysis of FS and three ensemble learning methods [6] combined FS and Data Balancing (DB) with ensemble techniques [7] proposed SmoteNDBoost and RusNDBoost [63] compared the bagging, boosting, and stacking ensembles. 11 base classifiers were used [64] RF was combined with feature selection and data sampling [65] analyze whether different classifiers identify the same defects or not using RF, NB, RPart, and SVM 2018 [8] analyzed ensembles of weighted randomized majority voting techniques [9] proposed SDAEsTSE model [12] analyzed performance model with and without applying SMOTE, AdaBoost, and Bagging [41] Proposed PBIL-Auto-Ens technique [42] proposed ensemble ROS, MWM, and FIDos methods with RF as base classifier [47] proposed multi-objective optimization for ensemble classification [52] proposed framework based on PCA in FS and RF, Adaboost, bagging, and classification via regression ensembles [55] proposed enhancement in the SMOTE-Ensemble approach using cost-sensitive learning (CSL) [66] proposed deep super learner (DSL) 2019 [67] compared Adaboost, Bagging, RSM, RF, and Vote ensembles [68] ten ensemble classifiers were compared to baseline classifiers 2020…”

Section: Rq1 : Which Ensemble Learning Techniques Are Applied For Software Defect Prediction?mentioning

confidence: 99%

“…In [67], precision rate, f-measure and AUC were applied to evaluate the defect prediction model. [68] used F-measure and AUC only to evaluate performance. In [69], AUC and accuracy were employed as a measure to compare performance.…”

Section: Rq2 : Which Evaluation Criterion Is Used To Measure the Performance Of Ensemble Learners?mentioning

confidence: 99%

“…Fig 5 shows the distribution of primary studies over machine learning tools and libraries. In our selected primary studies, we determined that Waikato Environment for Knowledge Analysis (WEKA) is the most frequently utilized machine learning tool to explore, visualize and perform classification over data [1], [3][4], [6], [8][9], [11][12], [41], [43], [45], [50][51][52][53][54][55], [58], [60], [62][63][64], [67][68], [73]. Other tools include MATLAB [9], [11], [47], [48] and sklearn [7], which is a python machine learning library and LIBSVM [56].…”

Section: Rq3 : What Are the Most Commonly Employed Tools To Implement Ensemble Learning Techniques?mentioning

confidence: 99%

See 3 more Smart Citations

Software Defect Prediction Using Ensemble Learning: A Systematic Literature Review

et al. 2021

View full text Add to dashboard Cite

Recent advances in the domain of software defect prediction (SDP) include the integration of multiple classification techniques to create an ensemble or hybrid approach. This technique was introduced to improve the prediction performance by overcoming the limitations of any single classification technique. This research provides a systematic literature review on the use of the ensemble learning approach for software defect prediction. The review is conducted after critically analyzing research papers published since 2012 in four well-known online libraries: ACM, IEEE, Springer Link, and Science Direct. In this study, five research questions that cover the different aspects of research progress on the use of ensemble learning for software defect prediction are addressed. To extract the answers to identified questions, 46 most relevant papers are shortlisted after a thorough systematic research process. This study will provide compact information regarding the latest trends and advances in ensemble learning for software defect prediction and provide a baseline for future innovations and further reviews. Through our study, we discovered that frequently employed ensemble methods by researchers are the random forest, boosting, and bagging. Less frequently employed methods include stacking, voting and Extra Trees. Researchers proposed many promising frameworks, such as EMKCA, SMOTE-Ensemble, MKEL, SDAEsTSE, TLEL, and LRCR, using ensemble learning methods. The AUC, accuracy, F-measure, Recall, Precision, and MCC were mostly utilized to measure the prediction performance of models. WEKA was widely adopted as a platform for machine learning. Many researchers showed through empirical analysis that feature selection and data sampling were important pre-processing steps that improve the performance of ensemble classifiers.

show abstract

Section: Rq1 : Which Ensemble Learning Techniques Are Applied For Software Defect Prediction?mentioning

confidence: 99%

Section: Rq1 : Which Ensemble Learning Techniques Are Applied For Software Defect Prediction?mentioning

confidence: 99%

Section: Rq2 : Which Evaluation Criterion Is Used To Measure the Performance Of Ensemble Learners?mentioning

confidence: 99%

Section: Rq3 : What Are the Most Commonly Employed Tools To Implement Ensemble Learning Techniques?mentioning

confidence: 99%

See 2 more Smart Citations

Software Defect Prediction Using Ensemble Learning: A Systematic Literature Review

et al. 2021

View full text Add to dashboard Cite

show abstract

“…e final results showed that, DP-CNN improves the state-of-the-art method by 12%. Yucalar et al [23], in their study, aimed at empirical demonstration of performance of fault prediction of 10 ensemble predictors with baseline predictor. e experiment was conducted on 15 open-source project datasets from PROMISE repository.…”

Section: Related Workmentioning

confidence: 99%

Impact of Parameter Tuning for Optimizing Deep Neural Network Models for Predicting Software Faults

Gupta

Rajnish

Bhattacherjee

2021

Scientific Programming

View full text Add to dashboard Cite

Deep neural network models built by the appropriate design decisions are crucial to obtain the desired classifier performance. This is especially desired when predicting fault proneness of software modules. When correctly identified, this could help in reducing the testing cost by directing the efforts more towards the modules identified to be fault prone. To be able to build an efficient deep neural network model, it is important that the parameters such as number of hidden layers, number of nodes in each layer, and training details such as learning rate and regularization methods be investigated in detail. The objective of this paper is to show the importance of hyperparameter tuning in developing efficient deep neural network models for predicting fault proneness of software modules and to compare the results with other machine learning algorithms. It is shown that the proposed model outperforms the other algorithms in most cases.

show abstract

MLPNN‐RF: Software fault prediction based on robust weight based optimization and Jacobian adaptive neural network

Babu¹,

Sivasubramanian

Natarajan

2022

Concurrency and Computation

View full text Add to dashboard Cite

Software fault prediction (SFP) is a vital objective in software engineering. This might permit effective resource allocation and also enhance informed decisions about the release quality. SFP is being a critical issue for software professionals as well as the tech industry. Thus, SFP is necessary. The study intends to perform efficient error rate estimation using the proposed hybrid robust weight based optimization and Jacobian adaptive neural network (RWO-JANN). It also aims to classify the software faults in an efficient way using multi-layer perceptron neural network-random forest (MLPNN-RF). Various processes are involved to accomplish SFP. At first, the dataset is taken as input. After this, data preprocessing is performed. Subsequently, weights are initialized using the proposed RWO-JANN. Weight initialization is performed through a series of steps. Then, the position and the weight parameter are updated to perform weight initialization. After this, the error rate is estimated and the weight is updated on the basis of the learning rate and Jacobian matrix calculation. Lastly, the decay rate is analyzed. If the error rate extends beyond the threshold value, the process repeats from weight initialization. If not, the testing process is performed and lastly the classified output for SFP is obtained by the proposed MLPNN-RF. The proposed system is comparatively analyzed with the existing methods in terms of accuracy, precision, recall, F1 score, sensitivity, specificity, and error rate. The analytical results revealed effective outcomes of proposed system than the existing techniques with accuracy of 99.01%.

show abstract

Multiple-classifiers in software quality engineering: Combining predictors to improve software fault prediction ability

Cited by 38 publications

References 29 publications

Software Defect Prediction Using Ensemble Learning: A Systematic Literature Review

Software Defect Prediction Using Ensemble Learning: A Systematic Literature Review

Impact of Parameter Tuning for Optimizing Deep Neural Network Models for Predicting Software Faults

MLPNN‐RF: Software fault prediction based on robust weight based optimization and Jacobian adaptive neural network

Contact Info

Product

Resources

About