A fuzzification algorithm,which is based on the structure adaptive tunning nerual network,is put forward in this paper.
从避免人工干预的思想出发,提出了一种基于结构自调整神经网络的模糊划分算法。
参考来源 - 用神经网络方法进行模糊划分设计The paper forward Nerual Network forecasting system based on grey correlation analysis. The way breaks a new path to further definition of dynamic stock market.
最后介绍了神经网络理论知识和BP网络学习算法和预测步骤,提出了基于灰色关联分析的神经网络预测系统,该系统为进一步明确描述动态的股市行情开辟了新的思路。
参考来源 - 基于灰色理论和神经网络理论的股票指数预测研究·2,447,543篇论文数据,部分数据来源于NoteExpress
Method:The main effective factors could be found by weights of the BP Nerual Network, which has one hidden layer, and the correlation of input and output.
方法:拟合输入与输出之间含有一个隐层的BP神经网络,利用各层输入与输出间的相关程度与网络权值确定各因素影响力大小。
Methods the main effective factors could be found by weights of the BP Nerual Network, which has one hidden layer, and the correlation of inputs and output.
方法拟合输入与输出之间含有一个隐层的BP神经网络,利用各层输入与输出间的相关程度与网络权值确定各因素影响力大小。
Using the reconfigurable system based on FPGA controlled by CPLD in the reconfigurable implementation of BP train Network and BP implement Network of Artificial Nerual Network.
并将由CPLD控制的基于FPGA的可重构系统应用到人工神经网络的BP网络训练和BP网络执行两个算法的可重构实现过程中并进行分析。
应用推荐