Neuro-genetic approach on logistic model based software reliability prediction

Roy, Pratik; Mahapatra, G. S.; Dey, Kashi Nath

doi:10.1016/j.eswa.2015.01.043

Cited by 28 publications

(12 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…NPSO is a scholastic optimization method which is successfully applied to train RNN. In this study, conventional NN is used to modify the network parameter and precision ([56], [23]) to improve the ability of the network.…”

Section: Main Textmentioning

confidence: 99%

“…The proposed model is trained by proposed neighborhood based PSO to find out the optimal solution ([56], [32]). The weight and parameters of the PNSRNN are considered by [normalw1,normalw2,normalw3,normalw40em0em′,normalw50em0em′,normalw60em0em′,p,q,r].…”

Section: Main Textmentioning

confidence: 99%

“…Here, the fitness function Nrmse is described for the purpose, which is given by Normalized Root Mean Square Error (NRMSE) measurement for minimization of the error generated by the RNN as follows:Nrmsebadbreakafter=truefalse∑i=1ntrue(yi′badbreakafter−yitrue)2truefalse∑i=1nyi2 where n denotes data points are applied to train the NN; yi and yi′ represent the actual and predicted value for the i -th software failure data point. NRMSE value is minimized by the proposed PSO algorithm during learning of the RNN [56].

…”

Section: Main Textmentioning

confidence: 99%

“…In this section, the performance of proposed model is compared with NN and PSO based DWCM with the Yamada Model [6], PNZ Model [7] and Roy Model [56] in software reliability modeling and analysis. We have exhibited the performance of the models using Goodness of fit, convergence by fitness function, End point prediction, and RPE metrics.…”

Section: Main Textmentioning

confidence: 99%

See 3 more Smart Citations

A novel approach of NPSO on dynamic weighted NHPP model for software reliability analysis with additional fault introduction parameter

Rani

Mahapatra²

2019

Heliyon

Self Cite

View full text Add to dashboard Cite

This paper presents software fault detection, which is dependent upon the effectiveness of the testing and debugging team. A more skilled testing team can achieve higher rates of debugging success, and thereby removing a larger fraction of faults identified without introducing additional faults. A complex software is often subject to two or more stages of testing that exhibits distinct rates of fault discovery. This paper proposes a two-stage Enhanced neighborhood-based particle swarm optimization (NPSO) technique with the assimilation of the three conventional non homogeneous Poisson process (NHPP) based growth models of software reliability by introducing an additional fault introduction parameter. The proposed neuro and swarm recurrent neural network model is compared with similar models, to demonstrate that in some cases the additional fault introduction parameter is appropriate. Both the theoretical and predictive measures of goodness of fit are used for demonstration using data sets through NPSO.

show abstract

Section: Main Textmentioning

confidence: 99%

Section: Main Textmentioning

confidence: 99%

…”

Section: Main Textmentioning

confidence: 99%

Section: Main Textmentioning

confidence: 99%

See 2 more Smart Citations

A novel approach of NPSO on dynamic weighted NHPP model for software reliability analysis with additional fault introduction parameter

Rani

Mahapatra²

2019

Heliyon

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, NN approaches are well capable to work on a universal approximator for any nonlinear continuous function with arbitrary accuracy . Karunanithi et al proposed ANN to predict the software reliability by using the execution time and the cumulative number of detected faults as the input and the desired output of the network, respectively: the cumulative execution time as inputs and the corresponding accumulated failures as desired outputs for NN approach for software reliability modelling . Tian and Noore presented an evolutionary NN‐based method for software reliability prediction using multiple delayed input—single‐output architecture and genetic algorithm to optimize the number of input nodes and hidden nodes of the NN.…”

Section: Introductionmentioning

confidence: 99%

Neural network for software reliability analysis of dynamically weighted NHPP growth models with imperfect debugging

Rani¹,

Mahapatra²

2018

Software Testing Verif & Rel

Self Cite

View full text Add to dashboard Cite

This paper propose a learning algorithm of supervised back-propagation neural networks for dynamic weighted combination of software reliability model. The proposed model is an assimilation of 3 well-known non-homogeneous poisson process (NHPP)-based software reliability growth models with imperfect debugging. The novel approach of proposed supervised back propagation-based neural network 2-stage architecture has a great impact on the network by combining the imperfect debugging models based on the nature of fault introduction rate during testing and debugging. Function approximation metrics are used for comparing the proposed model with individual models. Three data sets are trained using supervised back-propagation neural networks to compare the performance and validity evaluation of proposed and existing NHPP models and dynamic weighted combinational model. Reliability analysis among important NHPP models incorporating imperfect debugging is illustrated through numerical and graphical explanation of several metrics using supervised back-propagation neural networks. KEYWORDS imperfect debugging, fault introduction rate, neural networks, non-homogenous poisson process, software reliability growth model, supervised back-propagation algorithm INTRODUCTIONSoftware plays a vital role in every aspect of technological world with its ever-increasing dependency on technology, and its size and complexity have dramatically increased many folds [1]. Developers are facing a lot of problems to develop a trustworthy and reliable software for the user. A reliable software must meet or exceed the specified usage period, under specified operating conditions in a manner that satisfies customer's expectations.Software reliability is quantitative measurement of quality [2][3][4].Software cannot be developed error-free because of the uncertainties and amount of manual effort involved in the software development process. Specifically, erroneous software incurs high work costs. On the other hand, debugging and testing reduces the error content[5] but increases the development cost [6]. Software failure occurs because of hidden bugs in the software instead of immediate fixes for detected errors during development, but new errors may be introduced during debugging. Some reliability models have also been developed to address fault detection; correction [7, 8]; testing coverage [9]; and imperfect debugging processes [10, 11].Many factors such as software development methodology and complexity of the software are strongly dependant on making assumptions that influences the behaviour of software reliability growth. The software reliability measurement is based on type of data set [1]; therefore, traditional statistical theory plays a vital role in software reliability modelling. There are many constraints on existing statistical models that are based on prior assumptions in software testing. Neural network (NN) approaches can be applied to estimate and predict the software reliability. Neural network methods recommend far better models than ...

show abstract

Software reliability prediction: A machine learning and approximation Bayesian inference approach

Oveisi,

Moeini,

Mirzaei

et al. 2024

Quality & Reliability Eng

View full text Add to dashboard Cite

Reliability growth models are commonly categorized into two primary groups: parametric and non‐parametric models. Parametric models, known as Software Reliability Growth Models (SRGM) rely on a set of hypotheses that can potentially affect the accuracy of model predictions, while non‐parametric models (such as neural networks) can predict the model solely based on training data without any assumptions regarding the model itself. In this paper, we propose several methods to enhance prediction accuracy in software reliability context. More specifically, we, on one hand, introduce two gradient‐based techniques for estimating parameters of classical SRGMs. On the other, we propose methods involving LSTM Encoder–Decoder and Bayesian approximation within Langevin Gradient and Variational inference neural networks. To evaluate our proposed models' performance, we compare them with various neural network‐based software reliability models using three real‐world software failure datasets and utilizing the Mean Square Error (MSE) as a model comparison criterion. The experimental results indicate that our proposed non‐parametric models outperform most classical parametric and non‐parametric models.

show abstract

Neuro-genetic approach on logistic model based software reliability prediction

Cited by 28 publications

References 30 publications

A novel approach of NPSO on dynamic weighted NHPP model for software reliability analysis with additional fault introduction parameter

A novel approach of NPSO on dynamic weighted NHPP model for software reliability analysis with additional fault introduction parameter

Neural network for software reliability analysis of dynamically weighted NHPP growth models with imperfect debugging

Software reliability prediction: A machine learning and approximation Bayesian inference approach

Contact Info

Product

Resources

About