选择神经网络方法建立模型,模型的输入部分是经验证的64个风险因素,输出部分是软件项目的总体产出。首先验证模型输入输出部分的内容有效性,然后收集软件组织的实际软件项目样本用于建立和验证模型。实验结果表明,该模型能有效地测量软件项目的总体风险。
The model is based on neural network approach,and the input of model is 64 risk factors which have been verified,and the output of model is the total outcome of the project.We ensure the validity of the contents of input and output and collect software project samples to construct and test the model.We divide the samples into two parts,which are used for the training and the validity test of model respectively.It is proved that the model could evaluate the total risk of software project.