|
|||||
|
|
| 基于PCA-ANN模型中存在的问题及改进 | |
| 引用本文： | 朱俊峰. 基于PCA-ANN模型中存在的问题及改进[J]. 合肥学院学报(自然科学版), 2010, 20(3): 36-39 |
| 作者姓名： | 朱俊峰 |
| 作者单位： | 合肥工业大学,计算机与信息学院,合肥,230009;安徽广播电视大学,教务处,合肥,230022 |
| 摘 要： | 文章深入分析了目前普遍采用的主成分分析——神经网络模型应用中存在的不合理问题,通过推导指出错误所在,提出了相应的改进方案.为了验证改进模型的有效性,以UCI机器学习库中的数据集为样本,选取有导师BP神经网络和无导师SOM神经网络,建立改进的主成分分析——神经网络模型,并与传统主成分分析——神经网络模型进行比较测试,实验结果表明改进的模型效果更优. |
| 关 键 词： | 主成分分析 神经网络 算法 |
The Problem and Improvement of PCA-ANN Model |
|
| ZHU Jun-feng. The Problem and Improvement of PCA-ANN Model[J]. Journal of Hefei University(Natural Sciences Edition), 2010, 20(3): 36-39 | |
| Authors: | ZHU Jun-feng |
| Affiliation: | ZHU Jun-feng,(1.School of Computer and Information,Hefei University of Technology,Hefei 230009;2.Academic Affair Office,Anhui Radio and TV University,Hefei 230022,China) |
| Abstract: | This essay thorough analyze the principal component analysis which was widely used at present the application of Neural Networks is unreasonable.By inference,we know where the error is,and propose the reform program.In order to verify the validity of the improved model,we take UCI machine learning data as samples,and select the tutor BP Neural Network and the unsupervised SOM Neural Network to improve the principal component analysis the Neural Network model.Compared with the traditional principal component analysis and the Neural Network model,we can find that the improved model is better than the other. |
| Keywords: | principal component analysis neural network algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |