基于模糊神经网络的软件质量评价方法

为了对计算机软件质量进行更加精确、客观、科学的评价,综合模糊逻辑推理的结构性知识表达能力和神经网络的自学习能力,提出一种基于模糊神经网络的软件质量评价方法。首先筛选软件质量评价的指标建立评价等级模糊级,然后利用神经网络的学习能力,通过网络的训练样本,确定评价中的变量参数;训练好的网络就可用来对软件质量进行评价。应用表明,该方法克服了由专家确定变量参数的缺点,达到评价结果更高的客观性、科学性,具有很高的应用价值。

Method of Software Quality Evaluation Based on Fuzzy Neural Network

In order to make the software quality evaluation scientific and objective,integrating fuzzy logic theory and neural network theory,a method of software quality evaluation based on fuzzy-neuro network is put forward.First,select software quality evaluation factors,and set up the fuzzy sets of evaluation factors and scales,then the variables of evaluation are set by the study ability of neural network.The practical application shows that the method overcomes the defect that the variable set by experts,then makes the evaluation results objective and scientific.