Date Added: Oct 2012
Various models for software reliability prediction were proposed by many researchers. In this paper, the authors present a hybrid approach based on the Neural Networks and Simulated Annealing. An adaptive simulated Annealing algorithm is used to optimize the mean square of the error produced by training the Neural Network, predicting software cumulative failure. To evaluate the predictive capability of the proposed approach various projects were used. A comparison between this approach and others is presented. Numerical results show that both the goodness-of-fit and the next-step-predictability of their proposed approach have greater accuracy in predicting software cumulative failure compared with other approaches.