基于粗糙集和改进遗传算法优化BP神经网络的算法研究 -- 西北工业大学,2012,30(4):601-606
论文:2012,Vol:30,Issue(4):601-606
引用本文:
李伟, 何鹏举, 杨恒, 陈明. 基于粗糙集和改进遗传算法优化BP神经网络的算法研究[J]. 西北工业大学
Li Wei, He Pengju, Yang Heng, Chen Ming. An Effective Backpropagation Algorithm for Optimizing BP Neural Network Based on Rough Set and Modified Genetic Algorithm[J]. Northwestern polytechnical university

基于粗糙集和改进遗传算法优化BP神经网络的算法研究
李伟, 何鹏举, 杨恒, 陈明
1. 西北工业大学 自动化学院,陕西 西安 710072;
2. 无锡泛太科技有限公司,江苏 无锡 214000
摘要:
针对BP神经网络结构由于特征维数增多变得复杂,以及网络易陷入局部极值点,提出了粗糙集和改进遗传算法结合共同优化神经网络的方法。首先利用粗糙集对样本空间进行属性约简,降低特征维数,进而简化BP神经网络的结构;然后训练过程中先用改进的遗传算法全局搜索网络的权值和阀值,再使用BP算法局部搜索细化,避免网络过早收敛。试验分析证明优化后BP神经网络比传统BP网络的预测精度得到了极大提高,泛化能力得到了增强,说明了该方法的可行性、有效性。
关键词:    BP神经网络    粗糙集    遗传算法    属性约简    局部极值    权值和阀值   
An Effective Backpropagation Algorithm for Optimizing BP Neural Network Based on Rough Set and Modified Genetic Algorithm
Li Wei, He Pengju, Yang Heng, Chen Ming
1. Department of Automatic Control, Northwestern Polytechnic University, Xi'an 710072, China;
2. Wuxi Fantai Technology Co. , Ltd. , Wuxi 214000, China
Abstract:
Considering that the BP neural network became complex due to the increase of the sample dimension andit fell easily into local maximums or minimums,we combined genetic algorithm and rough set to optimize the BPneural network. Sections 1 through 3 explain our backpropagation algorithm mentioned in the title,which we be-lieve is effective and whose core consists of: (1) rough set was applied to simplify the network by reducing the at-tribute dimension; (2) modified genetic algorithm was used to globally search the weights and bios and,further,the BP algorithm was to locally optimize them to avoid the network falling into the local extremes. Simulation re-sults,presented in Fig. 1 and Table 2 in subsection 3. 4,and their analysis indicated preliminarily that predictionaccuracy was increased greatly over that of the traditional BP neural network and that generalization was enhanced,thus showing that our backpropagation algorithm is indeed effective.
Key words:    backpropagation algorithms    decision making    efficiency    errors    genetic algorithms    mathematicalmodels    neural networks    optimization    rough set theory;reduction of attribute dimension    simula-tion    weights and bios   
收稿日期: 2011-10-22     修回日期:
DOI:
基金项目: 陕西省科技攻关项目(2011K06-25)资助
通讯作者:     Email:
作者简介: 李伟(1980-),女,西北工业大学博士研究生,主要从事多传感器数据融合与模式识别的研究。
相关功能
PDF(386KB) Free
打印本文
把本文推荐给朋友
作者相关文章
李伟  在本刊中的所有文章
何鹏举  在本刊中的所有文章
杨恒  在本刊中的所有文章
陈明  在本刊中的所有文章

参考文献:
[1] 张东波, 黄辉先, 王耀南. 基于粗糙集约简的特征选择神经网络集成技术. 控制与决策, 2010, 25(3): 371-377Zhang Dongbo,Huang Huixian,Wang Yaonan. Feature Selection Neural Network Ensemble Based on Rough Sets Reducts.Control and Decision. 2010, 25(3): 371-377 (in Chinese)
[2] 张兆礼, 孙圣和. 粗神经网络及其在数据融合中的应用. 控制与决策, 2001, 16(1): 76-82Zhang Zhaoli,Sun Shenghe. Rough Neural Network and Its Application in Multisensor Data Fusion. Control and Decision.2001, 16(1):76-82 (in Chinese)
[3] 张 利,吴华玉,卢秀颖. 基于粗糙集的改进 BP 神经网络算法研究. 大连理工大学学报, 2009, 49(6): 971-976Zhang Li, Wu Huayu, Lu Xiuying. An Improved BP Neural Network Algorithm Based on Rough Set. Journal of Dalian Universi-ty of Technology. 2009, 49(6):971-976 (in Chinese)
[4] 苗夺谦, 李道国. 粗糙集理论、 算法与应用. 北京:清华大学出版社, 2008Miao Duoqian,Li Daoguo. Rough Set Theory,Algorithm and Application. Beijing: Tsinghua University Press,2008 (in Chinese)
[5] 张文修, 吴伟志, 梁吉业, 李德玉. 粗糙集理论与方法. 北京:科学出版社, 2001Zhang Wenxiu,Wu Weizhi,Liang Jiye,Li Deyu. Rough Set Theory and Methods. Beijing: Science Press, 2001 (in Chinese)
[6] Pawlak Z,et al. Rough Set Approach to Multi-Attribute Decision Analysis. European Journal of Operational Research,1994,72(3):443-459
[7] Pawlak Zdzislaw. Rough Set Approach to Knowledge-Based Decision Support. European Journal of Operational Research, 1997(99): 48-57
[8] Skowron A,Peters J. Rough Sets: Trends and Challenges. Rough Sets,Fuzzy Sets,Data Mining,and Granular Computing .Berlin: Springer-Verlag, 2003:25-34
相关文献:
1.郭阳明, 冉从宝, 姬昕禹, 马捷中.基于组合优化BP神经网络的模拟电路故障诊断[J]. 西北工业大学, 2013,31(1): 44-48