神经网络多指标综合评价方法研究

Study on The Multi -index Comprehensive Evaluation Method of Artificial Neural Network

利用人工神经网络BP算法建立了多指标综合评价模型,采用模糊数学的隶属函数对评价指标进行描述,由隶属函数的端点值和中间值组成学习样本模式,举例讨论了种经网络在多指标综合评价中的应用方法及应注意的一些问题.图1,表1,参5.

The multi-index comprehensive evaluation model was established by artificial neural network BP algorithm. The evaluation index was described by adopting subordinative function of fuzzy mathematics. Sample -study pattern consists of the values of subordinative function on ends and on middle point. An expmple was provided to discuss the application method of artificial neural network in the multi -index comprehensive evaluation and some notable problems.