计算机与现代化 ›› 2011, Vol. 1 ›› Issue (3): 12-14.doi: 10.3969/j.issn.1006-2475.2011.03.004
• 人工智能 • 上一篇 下一篇
黄文涛1，鲍 鸿1，张 晶2
收稿日期:
修回日期:
出版日期:
发布日期:
HUANG Wen-tao1, BAO Hong1, ZHANG Jing2
Received:
Revised:
Online:
Published:
摘要： 针对差异性是集成学习的一个重要条件,研究基于模糊聚类技术提高神经网络集成差异性的方法。提取大量弱分类器的权值和阈值并作为模糊聚类的数据对象，然后将聚类结果作为集成网络中个体网络的权值和阈值，最后在标准数据集上进行仿真实验，证实方法的有效性。
关键词: 差异性, 模糊聚类, 神经网络集成
Abstract: Aiming at diversity being an important condition of the ensemble learning, this paper studies the method for improving diversity of the neural networks ensemble based on fuzzy cluster algorithm. At first, a lot of weak classifiers weights and thresholds are collected as the original data of the fuzzy cluster algorithm. And then the individuals of the nearest clustering center are selected to make up of the memberships weights and thresholds of the ensemble learning. Finally, the experiment results prove the effectiveness of this method.
Key words: diversity, fuzzy cluster, neural networks ensemble
中图分类号:
TP391.4
黄文涛;鲍鸿;张晶. 基于模糊聚类的神经网络集成[J]. 计算机与现代化, 2011, 1(3): 12-14.
HUANG Wen-tao;BAO Hong;ZHANG Jing. Neural Networks Ensemble Based on Fuzzy Cluster[J]. Computer and Modernization, 2011, 1(3): 12-14.
0 / / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://www.c-a-m.org.cn/CN/10.3969/j.issn.1006-2475.2011.03.004
http://www.c-a-m.org.cn/CN/Y2011/V1/I3/12
