Performance Of Fault Prediction Models Research Articles

Empirical analysis of change metrics for software fault prediction

A quality assurance activity, known as software fault prediction, can reduce development costs and improve software quality. The objective of this study is to investigate change metrics in conjunction with code metrics to improve the performance of fault prediction models. Experimental studies are performed on different versions of Eclipse projects and change metrics are extracted from the GIT repositories. In addition to the existing change metrics, several new change metrics are defined and collected from the Eclipse project repository. Machine learning algorithms are applied in conjunction with the change and source code metrics to build fault prediction models. The classification model with new change metrics performs better than the models using existing change metrics. In this work, the experimental results demonstrate that change metrics have a positive impact on the performance of fault prediction models, and high-performance models can be built with several change metrics.

The Empirical Study of Semi-Supervised Deep Fuzzy C-Mean Clustering for Software Fault Prediction

Software fault prediction is a very consequent research topic for software quality assurance. The performance of fault prediction model depends on the features that are used to train it. Redundant and irrelevant features can hinder the performance of a classification model. In this paper, we propose an empirical study of two-stage data pre-processing technique on software fault prediction models. In the first stage, a novel semi-supervised deep Fuzzy C-Mean (DFCM) clustering-based feature extraction technique is proposed to create new features by utilizing deep multi-clusters of unlabeled and labeled data sets that tends to maximize intra-cluster class and intra-cluster feature by using FCM clustering. The FCM also utilizes to handle the class imbalance problem. In the second stage, we further ameliorate the prediction performance with coalescence of feature selection (using random-under sampling) to reduce the noisy data for classification. However, by the performance of the model results in the amalgamation of novel DFCM data pre-processing approach work better due to their ability to identify and amalgamation essential information in data features. An empirical study is designed on real-world software project (NASA & Eclipse) data set to evaluate the performance of DFCM by implemented different data pre-processing schemes on prediction models (C4.5, naive bayes, and 1-near neighbor (1-NN)), which are widely used in software fault prediction and further investigated the influencing factors in our approach. The result shows that the performance of the proposed DFCM feature extraction technique for data pre-processing is stable and effectiveness on all prediction models.

An in-depth study of the potentially confounding effect of class size in fault prediction

Background. The extent of the potentially confounding effect of class size in the fault prediction context is not clear, nor is the method to remove the potentially confounding effect, or the influence of this removal on the performance of fault-proneness prediction models.Objective. We aim to provide an in-depth understanding of the effect of class size on the true associations between object-oriented metrics and fault-proneness.Method. We first employ statistical methods to examine the extent of the potentially confounding effect of class size in the fault prediction context. After that, we propose a linear regression-based method to remove the potentially confounding effect. Finally, we empirically investigate whether this removal could improve the prediction performance of fault-proneness prediction models.Results. Based on open-source software systems, we found: (a) the confounding effect of class size on the associations between object-oriented metrics and fault-proneness in general exists; (b) the proposed linear regression-based method can effectively remove the confounding effect; and (c) after removing the confounding effect, the prediction performance of fault prediction models with respect to both ranking and classification can in general be significantly improved.Conclusion. We should remove the confounding effect of class size when building fault prediction models.

Comparative Analysis of Software Network and CK Metrics: Implications for Pre- and Post-release Faults

The research on open source software has attracted a great deal ofattention during the past decades for its wide applications in both academia and industry.Among the research topics related to open source software, the usefulness of software network metrics for faultprediction has beenreceived muchattention recently. In order to verifythe importance of software network metrics inthe performance of fault prediction models, this studymixes software network and Chidamber & Kemerer (CK)metrics to predict pre- and post-release fault-proneness of Eclipse version2.0. The small world and scale-free characteristics andSpearman correlation havebeen analyzed in detail. Inaddition,we employed10 classifiers, 2 weak classifiers, and 5 performanceevaluation criteria toconduct threesets of experiments. Experimental results have demonstrated that the software network metrics can only providelittle overall performanceimprovement.Based on the comprehensive study, the effect of software network metrics can be ignored.

Predict Fault-Proneness Module Using Pattern Recognization (A Review)

Background: The accurate prediction of where faults are likely to occur in code can help direct test effort, reduce costs and improve the quality of software. Objective of this paper is We investigate how the context of models, the independent variables used and the modeling techniques applied, influence the performance of fault prediction models. Method on We used a systematic literature review to identify 208 fault prediction studies published from January 2000 to December 2012. We synthesise the quantitative and qualitative results of 36 studies which report sufficient contextual and methodological information according to the criteria we develop and apply. Combinations of independent variables have been used by models that perform well. Feature selection has been applied to these combinations when models are performing particularly well. Conclusion: The methodology used to build models seems to be influential to predictive performance. Although

A Systematic Literature Review on Fault Prediction Performance in Software Engineering

Background: The accurate prediction of where faults are likely to occur in code can help direct test effort, reduce costs, and improve the quality of software. Objective: We investigate how the context of models, the independent variables used, and the modeling techniques applied influence the performance of fault prediction models. Method: We used a systematic literature review to identify 208 fault prediction studies published from January 2000 to December 2010. We synthesize the quantitative and qualitative results of 36 studies which report sufficient contextual and methodological information according to the criteria we develop and apply. Results: The models that perform well tend to be based on simple modeling techniques such as Naive Bayes or Logistic Regression. Combinations of independent variables have been used by models that perform well. Feature selection has been applied to these combinations when models are performing particularly well. Conclusion: The methodology used to build models seems to be influential to predictive performance. Although there are a set of fault prediction studies in which confidence is possible, more studies are needed that use a reliable methodology and which report their context, methodology, and performance comprehensively.

Thresholds based outlier detection approach for mining class outliers: An empirical case study on software measurement datasets

Predicting the fault-proneness labels of software program modules is an emerging software quality assurance activity and the quality of datasets collected from previous software version affects the performance of fault prediction models. In this paper, we propose an outlier detection approach using metrics thresholds and class labels to identify class outliers. We evaluate our approach on public NASA datasets from PROMISE repository. Experiments reveal that this novel outlier detection method improves the performance of robust software fault prediction models based on Naive Bayes and Random Forests machine learning algorithms.