基于级联神经网络组合模型的软件可靠性预测研究

Research of Software Reliability Prediction Based on Cascade Neural Network Combination Model

神经网络具有非常强的非线性计算能力,本文提出了一种新的级联神经网络模型,将BP神经网络模型和经典软件可靠性模型相融合形成一种新的前馈神经网络。这种前馈神经网络模型将传统的软件可靠性模型的输出作为BP神经网络的输入,逐级进行信息传递。

Neural network has a very strong nonlinear computing power. In this paper, a new cascade neural network model is proposed,which combines the BP neural network model with the classical software reliability model to form a new feedforward neural network. This feedforward neural network model uses the output of classical software reliability models as inputs to the BP neural network and information transfer step by step.