BP神经元网络算法在高新技术产品评价中的应用  

Application of BP Neural Network Algorithm in Evaluation of High-tech Products

作　　者：辛海涛[1] 杨殿军[1] 康靖[1] 

机构地区：[1]哈尔滨商业大学计算机与信息工程学院,黑龙江哈尔滨150028

出　　处：《商业经济》2010年第24期26-28,共3页Business & Economy

基　　金：黑龙江省自然基金资助(F2007-15);黑龙江省教育厅面上项目资助(11531065)

摘　　要：BP网络是一种按误差逆传播算法训练的多层前馈网络,是目前应用最广泛的神经网络模型之一。BP网络能学习和存贮大量的输入-输出模式映射关系,而无需事前揭示描述这种映射关系的数学方程。它的学习规则是使用最速下降法,通过反向传播来不断调整网络的权值和阈值,使网络的误差平方和最小。人工神经网络所具有的非线性特性、大量的并行分布结构以及学习和归纳能力使其在诸如建模、时间序列分析、模式识别、信号处理以及控制等方面得到广泛的应用。BP network is a feedforward neural network practiced by back propagation algorithm, is one of neural network modes applied widely. BP network is able to learn and store huge mapping relations of input and output modes, without diselesure and description of mathematical equation on the mapping relations beforehand. The rule is to take steepest descent method to adjust weight and threshold values constantly through antipropagation, make the nonlinear characteristic, huge parallel distribution, and learning and summing up a- bility of network square-error and the minimum artificial rural network used widely in modeling, analysis of time series, pattern recogni- tion, signal processing, control, and the like.

关 键 词：高新技术产品 人工神经元网络 应用评价 

分 类 号：F224.33[经济管理—国民经济]