基于BP算法的层次分析法研究

运用传统层次分析法进行系统评价与决策时，系统目标和评价指标的关系难以确定，而且评价过程中存在许多人为因素，影响评价的准确性和客观性，进而影响决策。基于BP算法的层次分析法利用BP神经网络的自学习、自适应、自组织功能，不断积累知识和经验，不断修正所学的知识，即修正网络神经元间的连接权值，通过给定的运算法则和激励函数，在输入评价指标后网络将输出最终的系统目标评价值，达到精确评价和最优决策的目的。

The Research of AHP Based on BP Arithmetic

When we evaluate the systems and decide which one is best,it is difficult to make sure of the relation of the system target and the evaluating targets using the traditional AHP,and there are many artificial factors in the course of evaluation,which affect the accuracy and objectivity of the evaluation,and then affect the decision.The AHP based on the BP arithmetic make use of the neural network's function of self-studying,self-adapting and self-organizing,constantly accumulates knowledge and experience,and revise its knowledge from learning continuously,namely revise the conjunction power of the neural network.The network will export the parameter of system's target after the input of evaluating target,by the giving algorithm and prompting function.With the operation output,we can attain the target of evaluation and decision accurately.