Software Reliability Prediction Based on Discrete Wavelet Transform and Neural Network

Effective prediction of the software reliability is one of the active pares of software engineering. This paper proposes a novel approach based on wavelet transform and neural network (NN). Using this approach, the time series of software faults can be decomposed into four components information, and then predict them by NN respectively. The experience results show that the performance of novel software reliability prediction approach is satisfactory.