A novel approach for software defect prediction through hybridizing gradual relational association rules with artificial neural networks

Miholca, Diana-Lucia; Czibula, Gabriela; Czibula, Istvan Gergely

doi:10.1016/j.ins.2018.02.027

Cited by 102 publications

(48 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance results of SGNE-PRRBC technique and existing methods [1] and [2] are discussed in this section with various metrics. With the assist of table and graphical representation, the performance is evaluated according to the following metrics.…”

Section: Resultsmentioning

confidence: 99%

“…These programs are taken from the school mate project files. The results show that the accuracy of SGNE-PRRBC technique is minimized as compared to the existing Fuzzy-filtered neuro-fuzzy framework [1] and HyGRAR [2] respectively. Figure 4 given above illustrates the accuracy of file classification.…”

Section: Impact Of Accuracymentioning

confidence: 98%

“…These weak learners" results are summed and obtain the strong classification results as an output. From the experimental, it is identified that for "5" program files are taken in the first iteration, the proposed SGNE-PRRBC technique correctly classified 4 program files and the accuracy is 80% whereas other two existing methods Fuzzy-filtered neuro-fuzzy framework [1] and HyGRAR [2] correctly classified 3 and 2 files and their accuracy are 60% and 40% respectively. The accuracy of SGNE-PRRBC technique is increased by 10% and 25% as compared to existing [1] and [2].…”

Section: Impact Of Accuracymentioning

confidence: 98%

“…However, the inter-version and inter-project fault prediction performance was not enhanced with lesser time. In [2], a gradual relational association rule mining and artificial neural networks (HyGRAR) was designed to discover the defective and non-defective software entities. The designed method failed to use an efficient machine learning technique for accurate defective prediction with minimum time.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Stochastic Embedded Probit Regressive Reweight Boost Classifier for Software Quality Examination

2019

IJRTE

View full text Add to dashboard Cite

In software development, Software quality analysis plays a considerable process. Through the software testing, the quality analysis is performed for efficient prediction of defects in the code. Due to the complicated structure of software projects, code examination has become a demanding issue that has to be addressed at the initial stage of testing for achieving the quality improved results. In order to resolve these issues, the Stochastic Gaussian Neighbor Embedding based Probit Regressive Reweight Boost Classification (SGNE-PRRBC) is introduced for accurate quality prediction system through code examination proficient system. The SGNE-PRRBC technique considers the number of program files as input for software quality analysis through feature selection and classification. Initially, the number of program files is taken from the dataset (DS). After collecting the files, the Gaussian distributive stochastic neighbor embedding technique choose the features (i.e. code metrics) based on the distance similarity. With the assist of Pearson correlative probit regressed reweight boost technique, the classification of program files is performed. The boosting algorithm creates ‘m’ number of weak classifiers i.e. Pearson correlative probit regression to categorize the input program files as normal or defected through analyze the source codes and chosen metrics. After that, the weak learners results are combined into strong through minimizing the out of sample error with gradient descent function. This enhances the accuracy of quality prediction and lessens the false positive rate (FPR). Experimental analysis is performed with various metrics namely accuracy, FPR and computation time (CT) with number of program files. Experimental results evident that the SGNE-PRRBC technique achieves better performance in terms of accuracy, CT and FPR as compared to the conventional methods.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Impact Of Accuracymentioning

confidence: 98%

Section: Impact Of Accuracymentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Stochastic Embedded Probit Regressive Reweight Boost Classifier for Software Quality Examination

2019

IJRTE

View full text Add to dashboard Cite

show abstract

“…Non-probability models, e.g. neural networks, have also been applied in the assessment of software reliability [11,12]. Another approach is software metrics assessment based on the defect per thousand line of codes (kilo line of code -KLOC).…”

Section: Introduction mentioning

confidence: 99%

A Bayesian Network Approach to Estimating Software Reliability of RSG-GAS Reactor Protection System

2019

View full text Add to dashboard Cite

Reliability represents one of the most important attributes of software quality. Assessing the reliability of software embedded in the safety of highly critical systems is essential. Unfortunately, there are many factors influencing software reliability that cannot be measured directly. Furthermore, the existing models and approaches for assessing software reliability have assumptions and limitations which are not directly acceptable for all systems, such as reactor protection systems. This paper presents the result of a study which aims to conduct quantitative assessment of the software reliability at the reactor protection system (RPS) of RSG-GAS based on software development life cycle. A Bayesian network (BN) is applied in this research and used to predict the software defect in the operation which represents the software reliability. The availability of operation failure data, characteristics of the RPS components and their operation features, prior knowledge on the software development and system reliability, as well as relevant finding from references were considered in the assessment and the construction of nodes on causal network model. The structure of causal model consists of eight nodes including design quality, problem complexity, and defect inserted in the software. The calculation result using Agenarisk software revealed that software defect in the operation of RPS follows binomial statistic distribution with the mean of 1.393. This number indicated the high software maturity level and high capability of the organization. The improvement of software defect concentration range on the posterior distribution compared with the prior's is also identified. The result achieved is valuable for further reliability estimation by introducing new evidence and experience data, and by setting up an appropriate plan in order to enhance software reliability in the RPS.

show abstract

A Novel Convolutional Neural Network Model to Predict Software Defects

Rajnish¹,

Bhattacherjee²,

Gupta³

2022

Fundamentals and Methods of Machine and Deep Learning

View full text Add to dashboard Cite

A novel approach for software defect prediction through hybridizing gradual relational association rules with artificial neural networks

Cited by 102 publications

References 25 publications

Stochastic Embedded Probit Regressive Reweight Boost Classifier for Software Quality Examination

Stochastic Embedded Probit Regressive Reweight Boost Classifier for Software Quality Examination

A Bayesian Network Approach to Estimating Software Reliability of RSG-GAS Reactor Protection System

A Novel Convolutional Neural Network Model to Predict Software Defects

Contact Info

Product

Resources

About