With the application of software systems in variety critical field the complexity level of software has increased, so software reliability has become more and more difficult to guarantee. To overcome human power and time limitation, researchers have focused on a soft computing approach. Nevertheless, the new techniques - especially NN - have some problems, like no solid mathematical foundation for analysis, trap in local minima and a convergence problem. This paper proposed a model to predict software reliability by hybridizing the Multi-Layer Perceptron neural network (MLP) and Imperialist Competitive algorithm (ICA). This model has solved most of the previous problems, such as the convergence problem, requiring a large amount of data, and it can be applied in complex systems. Numerical results show that both the training and testing stages of proposed approach have greater accuracy in predicting the number of software failures compared to the existing approaches.
CITATION STYLE
Noekhah, S., Salim, N. B., & Zakaria, N. H. (2018). Predicting software reliability with a novel neural network approach. In Lecture Notes on Data Engineering and Communications Technologies (Vol. 5, pp. 907–916). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-319-59427-9_93
Mendeley helps you to discover research relevant for your work.