|
|||||
|
|
| 一种新型的神经网络集成模型 | |
| 引用本文： | 张晓阳,徐敏,施化吉,李星毅.一种新型的神经网络集成模型[J].计算机工程与应用,2006,42(34):48-49,80. |
| 作者姓名： | 张晓阳 徐敏 施化吉 李星毅 |
| 作者单位： | 1. 江苏大学,梦溪图书馆,江苏,镇江,212013 2. 南通大学,计算机学院,江苏,南通,226007 3. 江苏大学,计算机科学与通讯工程学院,江苏,镇江,212013 |
| 基金项目： | 江苏省高校自然科学基金 |
| 摘 要： | 神经网络集成作为神经网络技术的延伸,被广泛的用于解决分类问题。很多实际应用表明:神经网络集成表现出比单个神经网络更好的性能。而传统的神经网络集成模型中网络的构建和集成是分两个阶段完成的。论文提出一种新的神经网络集成结构模型“层状集成”。该模型中网络的构建和集成同时完成,且每个成员网络的输出流入到下一个神经网络,作为下一个神经网络的输入,以这种方式生成一种层状神经网络集成。该模型用于解决分类问题,表现出比传统神经网络集成更好的性能。 |
| 关 键 词： | 神经网络 集成 层状模型 |
| 文章编号： | 1002-8331（2006）34-0048-02 |
| 收稿时间： | 2006-02 |
| 修稿时间： | 2006-02 |
New Module of Neural Network Ensembles |
|
| ZHANG Xiao-yang,XU Min,SHI Hua-ji,LI Xing-yi.New Module of Neural Network Ensembles[J].Computer Engineering and Applications,2006,42(34):48-49,80. | |
| Authors: | ZHANG Xiao-yang XU Min SHI Hua-ji LI Xing-yi |
| Affiliation: | 1.The Library of Jiangsu University,Zhenjiang,Jiangsu 212013,China; 2.School of Computer Science and Telecommunication Engineering,Jiangsu University, Zhenjiang,Jiangsu 212013,China; 3.College of Computer Science,Nantong University,Nantong, Jiangsu 226007, China |
| Abstract: | Neural network ensembles is widely used for classification problems as a development of the use of isolated networks.In many applications,ensembles has proven a performance above the performance of just one network.The standard model consists of two independent phases:model generation and model combination,and in this paper we present a new approach to neural network ensembles that we call "nappe ensembles".The approach is based on two ideas:the ensemble is created constructively,and the output of each network is fed to the inputs of the subsequent networks.In this way we make a cascade of networks.This method is compared with standard ensembles in solving several problems of classification with excellent performance. |
| Keywords: | neural network ensembles nappe ensemble |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |