论文部分内容阅读
神经网络作为解决当前大多数非线性科学和计算智能领域问题的主要工具,对解决一些定性问题定量处理发挥了重要的作用。标准化是介于社会科学和自然科学之间的一门复杂性科学,其具有非线性的典型特点,对一些指标进行量化就显得尤为困难,目前对一些单位和项目进行标准化评价,主要采用的方法是专家打分法,对相关的管理工作带来了一定的不便。本文借助于神经网络在处理非线性问题中的优势,通过标准化工作综合评价指标体系,建立了三层BP神经网络模型,并给出了相关的算法,从性能分析可以发现,该方法对于标准化工作的综合评价具有一定的先进性,对后续提升标准化工作的科学性具有一定的参考意义。
As the main tool to solve most current problems in the fields of nonlinear science and computational intelligence, neural networks play an important role in solving quantitative problems of some qualitative problems. Standardization is a complex science between social sciences and natural sciences. It has the characteristic of non-linearity. It is particularly difficult to quantify some indicators. At present, some units and projects are standardized and the main methods adopted Expert scoring method, the relevant management has brought some inconvenience. Based on the advantages of neural network in dealing with nonlinear problems, this paper establishes a three-layer BP neural network model through the standardization of comprehensive evaluation index system, and gives the relevant algorithms. From the performance analysis, we can find that this method is not suitable for standardization work Has a certain advanced nature of the comprehensive evaluation of the follow-up to enhance the scientific standardization of a certain reference value.