Pooja Rani
Posters-Accepted Abstracts: Adv Robot Autom
The main apparatus of soft computing paradigm are Artificial Neural Network, Fuzzy Logic, Swarm Intelligence and evolutionary computation. In this paper we introduce a new computing archetype present with soft computing tools and techniques to predict the software reliability concern with software developers and engineers in changing environment consideration. To construct a successful product we must have attributes including Reliability, performance, capability, functionality, usability, maintainability and documentation. Reliability is essentially being able to deliver usability of services while assuring the constraints of the system and is considered a part of quality assurance. The software reliability analysis is trait of software quality analysis. The objective of software reliability is also a very pertinent matter as per the customer satisfaction. To ensure models are developed to forecast the behaviour of software reliability. In order to achieve software reliability there are many reliability growth models are very complex and standard estimation procedures such as MLE (Maximum Likelihood Estimation) is difficult to estimate more realistic and useful so that the stochastic models over parametric models are very helpful for predicting. In this paper we investigate the potential benefits of using non-parametric modeling (NPM) methods to fit SRGMs (Software Reliability Growth Models) through soft computing techniques. The final numerically based example on real software failure data will be presented to illustrate the intelligence techniques developed and comparative work with some parametric SRGMs for software reliability prediction.
Advances in Robotics & Automation received 1275 citations as per Google Scholar report